WO2002019539A9 - Decodeur a sortie ponderee - Google Patents
Decodeur a sortie pondereeInfo
- Publication number
- WO2002019539A9 WO2002019539A9 PCT/JP2001/007576 JP0107576W WO0219539A9 WO 2002019539 A9 WO2002019539 A9 WO 2002019539A9 JP 0107576 W JP0107576 W JP 0107576W WO 0219539 A9 WO0219539 A9 WO 0219539A9
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- log likelihood
- decoding
- probability
- information
- soft output
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Classifications
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M13/00—Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes
- H03M13/29—Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes combining two or more codes or code structures, e.g. product codes, generalised product codes, concatenated codes, inner and outer codes
- H03M13/2957—Turbo codes and decoding
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M13/00—Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes
- H03M13/27—Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes using interleaving techniques
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M13/00—Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes
- H03M13/29—Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes combining two or more codes or code structures, e.g. product codes, generalised product codes, concatenated codes, inner and outer codes
- H03M13/2903—Methods and arrangements specifically for encoding, e.g. parallel encoding of a plurality of constituent codes
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M13/00—Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes
- H03M13/65—Purpose and implementation aspects
- H03M13/6566—Implementations concerning memory access contentions
Definitions
- the present invention relates to a soft output decoding device and a soft output decoding method for performing soft output decoding, and a decoding device and a decoding method suitable for iterative decoding.
- BACKGROUND ART In recent years, research has been conducted to reduce the symbol error rate by making the decoded output of an inner code in a concatenated code or the output of each iterative decoding operation in an iterative decoding method soft, to reduce the symbol error rate. Research on the law has been actively conducted.
- digital information is convolutionally coded by a coding device 1001 provided in a transmitting device (not shown), and the output is passed through a memoryless communication channel 1002 with noise. It is assumed that the data is input to a receiving device (not shown) via the receiving device, decoded by the decoding device 1003 provided in the receiving device, and observed.
- the transition probability P t (m I m,) between the states is defined by the following equation (1).
- P r ⁇ AIB ⁇ shown on the right side of the above equation (1) is the conditional probability that A occurs under the condition that B occurs.
- the no-memory storage channel 1002 with noise receives Y and outputs Y.
- the transition probabilities of the no-memory memory channel 1002 with noise are defined using the transition probabilities P r ⁇ yj IX of each symbol for all t (1 ⁇ t ⁇ T), as shown in the following equation (3). can do.
- Equation (4) represents the likelihood of the input information at time t when ⁇ is received, and is the soft output that should be originally obtained.
- the soft output can be expressed as the following equation (8).
- the decoding apparatus 1003 when performing the soft output decoding by applying the BCJR algorithm, the decoding apparatus 1003 obtains the soft output t by going through a series of steps shown in FIG. 3 based on these relationships.
- step S1001 the decoding device 1003 uses the above equations (9) and (11) to obtain a probability t (m) every time yt is received. ⁇ beauty ⁇ t (m ', m) is calculated.
- step S1002 the decoding device 1003 determines that all the sequences ⁇ ⁇ Then, the probability /? T (m) is calculated for each state m at all times t using the above equation (10).
- step S 1003 The decoder 1003 in step S 1003, step S 10 01, and step probability calculated in S 1002 shed t,? T and substituting into the above equation (8) ⁇ t, soft-output human t at each time t Is calculated.
- the decoding device 1003 can perform soft output decoding to which the BCJR algorithm is applied by going through such a series of processing.
- Max— L 0 g—BCJR algorithm expresses the probabilities t , / 5 t , ⁇ ”, and the soft output person t using logarithmic logarithm, using the product of probabilities as shown in the following equation (1 2).
- the probability sum operation is approximated by the logarithm maximum value operation as shown in the following equation (13).
- Y) is a function that selects the one with the larger value of x and y.
- I a t im sgn-log (a f ⁇ » ⁇ ))
- I f (w) sgn-log ( t (m))
- the decoding device 1003 can process both positive and negative values, so the constant sgn is either "+1" or "1-1". However, when the decoding device 1003 is configured as hardware, it is preferable to invert the negative sign of the calculated negative value and treat it as a positive value in order to reduce the number of bits. Is desirable.
- the decoding device 1003 is configured as a system that handles only negative values as the log likelihood, the constant sgn takes "+1" and the decoding device 1003 assumes the log likelihood. If the system is configured to handle only positive values, take "1-1". In the following, an algorithm considering such a constant s gn will be described.
- the decoding apparatus 1003 performs soft processing based on these relations through a series of steps shown in FIG. Find the output.
- the decoding device 1003 uses the above equations (15) and (17) every time y t is received in step S1011, log-likelihood I at (m) and I ⁇ t (m,, m) is calculated.
- step S1012 after receiving all of the sequence ⁇ ⁇ ⁇ ⁇ ⁇ ⁇ , the decoding device 1003 uses the above equation (16) to calculate each state m at all times t. Calculate the log likelihood (m).
- the decoding device 1 0 0 3 in step S 1 0 1 3, Step S 1 0 1 1 and Step S 1 0 1 Facial log likelihood I calculated at 2 t, I? T and I top ⁇ formula Substituting into (18), the log soft output I at each time t is calculated.
- the decoding device 1003 can perform soft output decoding to which the Max-L0g-BJR algorithm is applied by going through such a series of processing.
- the amount of operation can be significantly reduced as compared with the BJJ algorithm.
- the Log-BCJR algorithm improves the approximation accuracy of the Ma X-L0g-BCJR algorithm. Specifically, the Log-BCJR algorithm transforms the sum of probabilities shown in the above equation (13) by adding a correction term as shown in the following equation (19), This is to find the exact log value.
- the correction shall be referred to as 1 og-sum correction
- the operation shown on the left side of the above equation (19) is referred to as a 10 g-sum operation, and the operator of the 10 g-sum operation is referred to as "SS Pietrobon,” “Implemn tation and performance of a turbo / MAP decoder ", Int. J. Satellite Co-band un., vol. 16, pp. 23-46, Jan. -Feb. 1998". ), It is expressed as "#" (however, "E” in this paper) for convenience.
- the log likelihood I at, I 5 t and the log soft output I person t in the L 0 g—BCJR algorithm are given by the following equations (24) to (26), respectively. Can be represented as shown. Since the log likelihood Iat is represented by the above equation (17), its description is omitted here.
- the cumulative addition operation of the 10 g-sum operation in the right-hand state m, in the above equation (24) is obtained in the state m 'where the transition to the state m exists, and the above equation (25)
- the accumulative addition operation of the 10 g-sum operation in the state m on the right side of in shall be determined in the state m in which there is a transition from state m.
- the cumulative addition operation of the 10 g—sum operation of the first term on the right side in the above equation (26) is performed in the state m, where the transition to the state m exists when the input is “1”.
- the cumulative addition operation of the log-sum operation in the second term shall be obtained in state m 'where there is a transition to state m when the input is "0". Therefore, when soft output decoding is performed by applying the Log-BCJR algorithm, the decoding apparatus 1003 obtains a soft output based on these relations by going through the series of steps shown in FIG. be able to.
- the decoding apparatus 1003 uses the above equations (24) and (17) to obtain the log likelihood I at every time y t is received in step S1011. (m) and I ⁇ t (m,, m) is calculated.
- step S1012 after receiving all of the sequence Y, the decoding apparatus 1003 uses the above equation (25) to calculate the log likelihood I? Calculate t (m).
- step S1013 the decoding apparatus 1003 substitutes the log likelihood I / I / 5t and Iat calculated in step S1011 and step S1012 into the above equation (26), The log soft output I At at each time t is calculated.
- the decoding apparatus 1003 can perform soft output decoding to which the Log-B CJR algorithm is applied by going through such a series of processing.
- ROM Read Only Memory
- Such a Log-BCJR algorithm increases the amount of computation compared to the Max-Log-B CJR algorithm, but does not include the product operation, and its output is the same as that of the BCJR algorithm except for the quantization error. It is nothing but the logarithmic value of the soft output itself.
- the above-mentioned BCJR algorithm, Max—L 0 g—BCJR algorithm or L 0 g—BCJR algorithm is an algorithm that enables decoding of a trellis code such as a convolutional code.
- the present invention can also be applied to decoding of a code generated by connecting a plurality of element encoders via an interleaver. That is, the BCJR algorithm, the Max-L0g-BCJR algorithm or the L0g-BCJR algorithm are parallel concatenated convolutional codes (PCCC) or cascade concatenated convolutional codes (PCCC).
- SCCC Turbo-coded modulation
- PCCC or SCCC Turbo-coded modulation
- TTCM Turbo Trellis Coded Modulation
- SCTCM tandem concatenated Trellis Coded Modulation
- the decoding device that decodes these PC CC, S CCC, TT CM or SCT CM uses a maximum posterior probability (MAP) based on the BCJR algorithm, Max-Log-BCJ R algorithm, or LogB CJR algorithm. So-called iterative decoding is performed among a plurality of decoders that perform decoding.
- MAP maximum posterior probability
- each decoder when the code is punctured, the information necessary for performing the soft output decoding indicates a position where there is no code output such as a punctured position. Erase location information needs to be entered. This erase position The information must be separately stored at least until the log likelihood I is calculated.Therefore, each decoder must take measures such as providing a storage circuit that holds this erasure position information. However, this was a factor that put pressure on the circuit scale.
- each decoder decodes the parallel path as two paths even though the transition source state is the same and the transition destination state is the same, and can be simulated as one path. There is a need. In particular, in each decoder, the number of paths to be considered when calculating the log likelihood I, 1? Increases, so the processing load becomes large.
- An object of the present invention is to provide a soft-output decoding device and a soft-output decoding method that can reduce the processing load and increase the processing speed without deteriorating performance. It is another object of the present invention to provide a decoding apparatus and a decoding method suitable for iterative decoding that can reduce the processing load and increase the processing speed without deteriorating performance.
- An object of the present invention is to provide a soft-output decoding device and a soft-output decoding method capable of decoding with a simple configuration having a small circuit scale. Also, the present invention provides a decoding device suitable for iterative decoding that can decode with a simple configuration having a small circuit scale, and It is an object to provide a decoding method.
- the soft output decoding apparatus and method according to the present invention obtains a log likelihood that expresses the probability of passing through an arbitrary state based on a received value that is a soft input.
- the soft output decoding apparatus and method according to the present invention that achieves the above object obtains a log likelihood in which the probability of passing an arbitrary state is logarithmically expressed based on a received value that is a soft input
- a soft output decoding apparatus and method for decoding using this log likelihood comprising: a first log likelihood logarithmic representation of a first probability determined by a code output pattern and a received value for each received value.
- First probability calculating means for calculating the degree
- first probability distributing means for distributing the first log likelihood so as to correspond to a branch on a trellis according to a code configuration, and a state of a transition source
- a parallel path processing means for bundling the first log likelihood corresponding to the parallel path when decoding a code including a parallel path having the same and the same transition destination state.
- the soft output decoding apparatus and method according to the present invention obtains a log likelihood that expresses the probability of passing through an arbitrary state based on a received value that is a soft input.
- a soft output decoding apparatus and method for decoding using likelihood comprising: for each received value, a first log likelihood logarithmically expressing a first probability determined by a code output pattern and the received value.
- a second probability calculating means for calculating a second log likelihood in logarithmic notation of the probability of 2 and respective states in the reverse order of time series from the truncated state for each received value based on the first log likelihood.
- a third probability calculating means for calculating a third log likelihood in logarithmic notation of a third probability leading to, a first log likelihood, a second log likelihood, and a third log likelihood
- a soft output calculating means for calculating a log soft output in which the soft output at each time is logarithmically expressed, wherein the second probability calculating means calculates a second log likelihood in a process of calculating a second log likelihood.
- Two The sum of the log likelihood and the first log likelihood is supplied to a soft output calculating means.
- FIG. 2 is a diagram for explaining a trellis in a conventional encoding apparatus, and is a diagram for explaining the contents of probabilities,? And?.
- FIG. 3 is a flowchart illustrating a series of steps in performing a soft output decoding by applying a BJR algorithm in a conventional decoding device.
- FIG. 4 is a flowchart illustrating a series of steps in performing a soft-output decoding by applying a Max-Log-BCJR algorithm in a conventional decoding device.
- FIG. 5 is a block diagram illustrating a configuration of a communication model to which the data transmission / reception system shown as an embodiment of the present invention is applied.
- FIG. 6 is a block diagram illustrating a configuration of an example of an encoding device in the data transmission / reception system, and is a block diagram illustrating a configuration of an encoding device that performs encoding by PCC.
- FIG. 7 is a block diagram illustrating a configuration of an example of a decoding device in the data transmission / reception system, and is a block diagram illustrating a configuration of a decoding device that decodes a code by the encoding device illustrated in FIG. It is.
- FIG. 8 is a block diagram illustrating a configuration of an example of an encoding device in the data transmission / reception system, and is a block diagram illustrating a configuration of an encoding device that performs SCCCC encoding.
- FIG. 9 is a block diagram illustrating a configuration of an example of a decoding device in the data transmission / reception system, and is a block diagram illustrating a configuration of a decoding device that performs code decoding by the encoding device illustrated in FIG. It is.
- FIG. 10 is a block diagram illustrating a schematic configuration of the element decoder.
- FIG. 11 is a block diagram illustrating a detailed configuration of a left half portion of the same element decoder.
- C is a block diagram illustrating a detailed configuration of a right half portion of the same element decoder c.
- FIG. 13 is a block diagram illustrating a configuration of a decoded reception value selection circuit provided in the same element decoder.
- FIG. 14 is a block diagram illustrating a configuration of an edge detection circuit provided in the element decoder.
- FIG. 15 is a block diagram illustrating a schematic configuration of a soft output decoding circuit included in the element decoder.
- FIG. 16 is a block diagram illustrating the detailed configuration of the left half of the soft output decoding circuit.
- FIG. 17 is a block diagram illustrating a detailed configuration of a right half portion of the soft output decoding circuit.
- FIG. 18 is a block diagram illustrating a configuration example of a Posencraft type convolutional encoder.
- FIG. 19 is a block diagram illustrating another configuration example of the Posencraft type convolutional encoder.
- FIG. 20 is a block diagram illustrating a configuration example of a Massy type convolutional encoder.
- FIG. 21 is a block diagram illustrating another configuration example of the Massy type convolutional encoder.
- FIG. 22 is a block diagram illustrating a specific configuration example of the convolutional encoder illustrated in FIG.
- FIG. 23 is a diagram illustrating a trellis in the convolutional encoder shown in FIG.
- FIG. 24 is a professional / successful diagram for explaining a specific configuration example of the convolutional encoder shown in FIG.
- FIG. 25 is a diagram illustrating a trellis in the convolutional encoder shown in FIG.
- FIG. 26 is a block diagram illustrating a specific configuration example of the convolutional encoder illustrated in FIG.
- FIG. 27 is a diagram illustrating a trellis in the convolutional encoder shown in FIG. You.
- FIG. 28 is a block diagram illustrating a specific configuration example of the convolutional encoder illustrated in FIG.
- FIG. 29 is a diagram illustrating a trellis in the convolutional encoder shown in FIG.
- FIG. 30 is a block diagram illustrating a configuration of an internal erasure information generation circuit included in the soft output decoding circuit.
- FIG. 31 is a block diagram illustrating a configuration of a termination information generation circuit included in the soft output decoding circuit.
- FIG. 32 is a block diagram illustrating a configuration of a received value and prior probability information selection circuit included in the soft output decoding circuit.
- FIG. 33 is a block diagram illustrating a configuration of an Ia calculation circuit included in the soft output decoding circuit.
- FIG. 34 is a block diagram illustrating a configuration of an I y distribution circuit included in the soft output decoding circuit.
- FIG. 35 is a block diagram illustrating a configuration of a parallel path processing circuit for I ⁇ 0 included in the Ia distribution circuit.
- FIG. 36 is a block diagram illustrating a configuration of a parallel path lug-sum operation circuit included in the I0 parallel path processing circuit.
- FIG. 37 is a block diagram illustrating a configuration of an I-line calculation circuit included in the soft-output decoding circuit.
- FIG. 38 is a block diagram illustrating the configuration of the addition / comparison / selection circuit included in the I-parameter calculation circuit, in which two paths arrive from each state on the trellis to the state at the next time.
- FIG. 4 is a block diagram illustrating a configuration of an addition / comparison / selection circuit that performs processing on various codes.
- FIG. 39 is a block diagram illustrating a configuration of a correction term calculation circuit included in the addition / comparison / selection circuit.
- FIG. 40 is a block diagram for explaining the configuration of the addition / comparison / selection circuit included in the I-parameter calculation circuit, and includes four lines from each state on the trellis to the state at the next time.
- FIG. 3 is a block diagram illustrating a configuration of an addition / comparison / selection circuit that performs processing on a code that reaches a path.
- FIG. 41 is a block diagram illustrating a configuration of an I ⁇ + I key calculation circuit included in the I-parameter calculation circuit.
- FIG. 42 is a block diagram illustrating a configuration of an I? Calculation circuit included in the soft output decoding circuit.
- FIG. 43 is a block diagram illustrating the configuration of the addition / comparison / selection circuit included in the I-calculation circuit, in which two paths arrive from each state on the trellis to the state at the next time.
- FIG. 2 is a block diagram illustrating a configuration of an addition / comparison / selection circuit that performs processing on various codes.
- Fig. 44 is a block diagram illustrating the configuration of the addition / comparison / selection circuit of the I? Calculation circuit, in which four paths arrive from each state on the trellis to the state at the next time.
- FIG. 6 is a block diagram illustrating a configuration of an addition / comparison / selection circuit that performs processing on various codes.
- FIG. 45 is a block diagram illustrating a configuration of a soft output calculation circuit included in the soft output decoding circuit.
- FIG. 46 is a block diagram illustrating a configuration of a 10 g-s Um operation circuit included in the soft output calculation circuit.
- FIG. 47 is a block diagram illustrating a configuration of a reception value or prior probability information separation circuit included in the soft output decoding circuit.
- FIG. 48 is a block diagram illustrating a configuration of an external information calculation circuit included in the soft output decoding circuit.
- FIG. 49 is a block diagram illustrating a configuration of a hard decision circuit included in the soft output decoding circuit.
- FIG. 50 is a block diagram for explaining the concept of RAM for delay in the interleaver provided in the same element decoder.
- FIG. 51 is a block diagram for explaining the concept of the delay RAM, and is a block diagram for explaining that it is composed of a plurality of RAMs.
- FIG. 52 is a block diagram for explaining the concept of the RAM for delay.
- FIG. 4 is a block diagram for explaining how addresses generated by a control circuit of the interpolator are appropriately converted and given to each RAM.
- FIG. 53 is a block diagram for explaining the concept of an interleaving RAM in the interleaver.
- Figure 54 is a block diagram for explaining the concept of the interleave RAM. Based on the sequential write address and random read address, it is converted to the addresses used for punctures A and B.
- FIG. 4 is a block diagram for explaining a state given to each RAM.
- FIG. 55A is a diagram for explaining random interleave for input data of one symbol performed by the interleaver.
- FIG. 55B is a diagram for explaining random-interleave for two-symbol input data performed by the receiver.
- FIG. 55C is a diagram for explaining an interleave operation for input data of two symbols performed by the interleaver.
- FIG. 55D is a diagram for explaining paired interleaving of two-symbol input data performed by the interleaver.
- FIG. 55E shows random interleaving for input data of 3 symbols performed by the interleaver
- F is a diagram for explaining inline interleaving for input data of 3 symbols.
- FIG. 55G is a diagram for explaining paired interleave for input data of three symbols performed by the interleaver.
- FIG. 56 is a block diagram illustrating the configuration of the interleaver.
- FIG. 57 is a block diagram illustrating a configuration of an odd-length delay compensation circuit included in the connector.
- FIG. 58 is a block diagram illustrating a configuration of a storage circuit included in the interleaver.
- FIGS. 59A to 59D are diagrams for explaining a method of using the RAM of the receiver when random interleaving is performed on one symbol of input data.
- Figure 59A shows the RAM for delay
- Figure 59B shows the RA for in-leave.
- M shows the RAM for the address
- FIG. 59D shows the unused RAM.
- FIGS. 60A to 60D are diagrams for explaining a method of using the RAM of the interleaver in a case where random in-leave is performed on input data of two symbols.
- Figure 60A shows the RAM for delay
- Figure 60B shows the RAM for in-leaves
- Figure 60C shows the RAM for addresses
- Figure 60D shows the unused RAM. It is.
- FIGS. 61A to 61C are diagrams for explaining a method of using the RAM of the input / output of the two symbols in the case of performing the inline interleaving on the input / output of the two symbols.
- FIG. 61A shows a RAM for delay
- FIG. 61B shows a RAM for all-in-leave
- FIG. 61C shows a RAM for addresses.
- FIGS. 62A to 62D are diagrams for explaining a method of using the RAM of the input / output of the same symbol when performing pairwise interleave on the input / output of the two symbols.
- Figure 62A shows the RAM for delay
- Figure 62B shows the RAM for interleaving
- Figure 62C shows the RAM for addresses
- Figure 62 shows the RAM not used. It is.
- FIGS. 63A to 63D are diagrams for explaining a method of using the RAM of the interleaver when random interleave is performed on input data of three symbols.
- Figure 63A shows RAM for delay
- Figure 63B shows RAM for interleaving
- Figure 63C shows RAM for addresses
- Figure 63D shows RAM not used. It is.
- FIGS. 64A to 64D are diagrams for explaining a method of using the RAM of the interleaver in the case of performing an in-line in-time relieving on an input data of three symbols.
- Figure 64A shows RAM for delay
- Figure 64B shows RAM for interleaving
- Figure 64C shows RAM for addresses
- Figure 64D shows RAM not used.
- FIGS. 65A to 65D are diagrams for explaining a method of using the RAM of the interval bar when performing pairwise interleaving on input data of 3 symbols.
- Figure 65A shows the RAM for the delay
- Figure 65B shows the RAM for the FIG. 65C shows a RAM for an address
- FIG. 65D shows a RAM not used.
- FIG. 66 is a block diagram illustrating a configuration of a decoding device configured by connecting the same element decoders.
- FIG. 67 is a block diagram illustrating a simplified configuration of two adjacent element decoders constituting the decoding apparatus. The information necessary for soft-output decoding is selected from information from the preceding element decoder.
- FIG. 4 is a block diagram illustrating a configuration for selecting information.
- FIG. 68 is a block diagram illustrating a simplified configuration of two adjacent element decoders constituting the decoding apparatus.
- the element decoder at the previous stage performs soft output decoding at the element decoder at the next stage.
- FIG. 4 is a block diagram illustrating a configuration for selecting necessary information.
- FIG. 69 is a block diagram illustrating a simplified configuration of two adjacent element decoders constituting the decoding device, and is a block diagram illustrating a configuration including a delay circuit that delays a received value. is there.
- FIG. 70 is a block diagram illustrating a simplified configuration of two adjacent elementary decoders constituting the decoding apparatus.
- the configuration includes a decoded reception value selection circuit that selects a reception value to be decoded.
- FIG. 3 is a block diagram for explaining the configuration of FIG.
- FIGS. 71A to D are diagrams for explaining trellis in the convolutional encoder shown in FIG. 18 and explaining numbering based on a branch input as viewed from a state of a transition destination
- FIG. Figure 71A shows the numbering when the number of memories is four
- Figure 71B shows the numbering when the number of memories is three
- Figure 71C shows the numbering when the number of memories is two
- FIG. 71D is a diagram showing the numbering when the number of memories is one.
- FIGS. 72A to D are diagrams for explaining the trellis in the convolutional encoder shown in FIG. 18 and explaining the numbering based on the branch output from the transition source state.
- FIG. Fig. 72A shows the numbering when the number of memories is four
- Fig. 72B shows the numbering when the number of memories is three
- Fig. 72C shows the numbering when the number of memories is two.
- FIG. 72D is a diagram showing the numbering when the number of memories is one.
- FIGS. 73A and B are diagrams for explaining the trellis in the convolutional encoder shown in FIG. 19, and explain the numbering based on the input branch viewed from the transition destination state.
- FIG. FIG. 73A shows the numbering when the number of memories is three
- FIG. 73B shows the numbering when the number of memories is two.
- FIGS. 74A and B illustrate the trellis in the convolutional encoder shown in FIG. 19, and illustrate the numbering based on the branch output from the state of the transition source.
- FIG. FIG. 74A shows the numbering when the number of memories is three
- FIG. 74B shows the numbering when the number of memories is two.
- FIGS. 75A and 75B are diagrams for explaining the trellis in the convolutional encoder shown in FIG. 20 and explaining the numbering based on a branch inputted from the state of the transition destination
- FIG. 75A shows the numbering when the number of memories is three
- FIG. 75B shows the numbering when the number of memories is two.
- FIGS. 76A and 76B are diagrams for explaining the trellis in the convolutional encoder shown in FIG. 20 and explain the numbering based on the branch output from the state of the transition source.
- FIG. FIG. 76A shows the numbering when the number of memories is three
- FIG. 76B shows the numbering when the number of memories is two.
- FIGS. 77A and B are diagrams for explaining the trellis in the convolutional encoder shown in FIG. 21 and explaining numbering based on a branch input as viewed from the state of the transition destination
- FIG. 77A shows the numbering when the number of memories is two
- FIG. 77B shows the numbering when the number of memories is one.
- FIGS. 78A and B are diagrams for explaining the trellis in the convolutional encoder shown in FIG. 21 and explain the numbering based on the branch output from the state of the transition source.
- FIG. FIG. 78A shows the numbering when the number of memories is two
- FIG. 78B shows the numbering when the number of memories is one.
- FIG. 79 is a diagram showing a trellis for explaining the operation of generating termination information, and is a diagram for explaining an operation of inputting termination information for the number of input bits for the termination period.
- FIG. 80 is a diagram illustrating a trellis for describing an operation of generating termination information, and is a diagram illustrating an operation of inputting termination information in one time slot.
- Fig. 81 is a block diagram illustrating the schematic configuration of the I-key output circuit and the I-key distribution circuit. The log-likelihood I for all the input / output patterns is calculated, and according to the code configuration.
- FIG. 9 is a block diagram illustrating a configuration for distributing according to a determined input / output pattern.
- FIG. 82 is a block diagram illustrating a schematic configuration of the Ia calculation circuit and the Ia distribution circuit.
- the log likelihood Ia for at least a part of the input / output patterns is extracted, and a desired log likelihood is obtained. It is a block diagram explaining the structure which selects and adds the degree I key.
- FIG. 83 is a block diagram illustrating a schematic configuration of the Ia calculation circuit and the Ia distribution circuit. In the case where the log likelihood I for all input / output patterns is calculated, the log likelihood I
- FIG. 3 is a block diagram illustrating a configuration for performing normalization for each time for each key.
- FIGS. 84A and 84B are diagrams for explaining the normalization of the log likelihood Ia when the same element decoder treats the log likelihood as a negative value.
- FIG. 84A shows an example of distribution of log likelihood Ia before normalization
- FIG. 84B shows an example of distribution of log likelihood Ia after normalization.
- FIGS. 85A and 85B are diagrams for explaining the normalization of the log likelihood Ia when the same element decoder treats the log likelihood as a positive value.
- FIG. 85A shows an example of distribution of log likelihood Ia before normalization
- FIG. 85B shows an example of distribution of log likelihood Ia after normalization.
- FIG. 86 is a block diagram illustrating a schematic configuration of the Ia calculation circuit and the Ia distribution circuit. In the case where the log likelihood Ia for at least a part of the input / output patterns is calculated, the log likelihood is calculated.
- FIG. 4 is a block diagram for explaining a configuration for performing normalization for each time for a degree I.
- FIGS. 87A to 87D are diagrams illustrating an example of a trellis in a convolutional encoder.
- FIG. 87A shows an example when the number of memories is “1”
- FIG. 87B shows an example when the number of memories is “2”
- FIG. 87C shows an example when the number of memories is “3”.
- FIG. 87D is a diagram showing an example in the case where the number of memories is “4”.
- FIG. 88 is a diagram for explaining a state in which the four trellis shown in FIG. 87 are overlapped.
- Figure 89 shows two paths from each state on the trellis to the state at the next time.
- FIG. 2 is a block diagram illustrating a configuration of an addition / comparison / selection circuit in the I / H calculation circuit that performs processing on a code that arrives at a location where a selector is provided for selecting a log likelihood I ⁇ . It is a block diagram.
- FIG. 90 is a block diagram for explaining a schematic configuration of a 10 g-sum calculation circuit in the I-calculation circuit and the I-calculation circuit, and performs 10 g-s normalization by the first method.
- FIG. 3 is a block diagram illustrating a configuration of a um operation circuit.
- FIG. 91 is a diagram for explaining the normalization by the first method, and is a diagram illustrating an example of a dynamic range before and after the normalization.
- FIG. 92 is a diagram for explaining the normalization by the second method, and is a diagram showing an example of the dynamic range before and after the normalization.
- FIG. 93 is a block diagram illustrating a schematic configuration of a 10 g-sum calculation circuit in the I-calculation circuit and the I-calculation circuit, and performs 10 g-s normalization by the third method.
- FIG. 3 is a block diagram illustrating a configuration of a um operation circuit.
- FIG. 94 is a diagram for explaining the normalization by the third method, and is a diagram illustrating an example of the dynamic range before and after the normalization.
- FIG. 95 is a block diagram illustrating a schematic configuration of a 10 g-sum arithmetic circuit, and is a block diagram illustrating a configuration of a 10 g-sum arithmetic circuit that performs a normal 1 og-sum arithmetic operation. is there.
- FIG. 96 is a block diagram illustrating a schematic configuration of a log-sum arithmetic circuit, in which values of a plurality of correction terms corresponding to a difference value are calculated, and an appropriate one is selected from the calculated values.
- FIG. 9 is a block diagram illustrating a configuration of a 10 g-sum operation circuit that performs um operation.
- FIG. 97 is a block diagram illustrating a schematic configuration of a soft output calculation circuit that performs a cumulative addition operation of a 1 og-sum operation without using an enable signal.
- FIGS. 98A to 98D are diagrams for explaining normalization of external information in symbol units.
- FIG. 98A shows an example of the distribution of external information before normalization
- FIG. 98B shows an example of the distribution of external information after normalization in which the external information having the maximum value is adjusted to a predetermined value
- FIG. Fig. 98D shows an example of the distribution of external information after clipping.
- Fig. 98D shows a positive difference in which the value of external information for one symbol is different from the value of external information for another symbol. It is a figure showing the example of distribution of external information after normalization.
- FIG. 99 is a diagram for explaining the signal point arrangement by the 8PSK modulation method, and is a diagram showing a state where a boundary line is provided on the I / Q plane.
- FIG. 100 is a block diagram illustrating a simplified configuration of a control circuit included in the connector.
- FIG. 101 is a diagram for explaining data write and read timings when a write address counter and a read address counter are shared.
- FIG. 102 is a diagram for explaining the write and read timings of the data when the counter for the write address and the counter for the read address are separately provided.
- FIG. 103 is a diagram for explaining how data is written to and read from the RAM in the interleaver.
- FIG. 104 is a diagram for explaining a state in which continuous addresses are assigned to RAMs in the same receiver.
- FIG. 105 is a diagram for explaining how data is written and read to and from the RAM in the same interleaver, and the data is stored when no data is stored in all storage areas of each RAM.
- FIG. 3 is a diagram for explaining a state of writing and reading.
- Fig. 106 is a diagram for explaining how continuous addresses are assigned to RAMs in the same server, and shows continuous addresses over a plurality of physically different RAMs. It is a figure for explaining a mode that is allocated.
- Fig. 107 is a diagram for explaining how addresses are allocated to RAMs in the same interleaver, and for explaining how a replacement destination address is given by a combination of a time slot and an input symbol.
- FIG. 107 is a diagram for explaining how addresses are allocated to RAMs in the same interleaver, and for explaining how a replacement destination address is given by a combination of a time slot and an input symbol.
- FIG. 108 is a diagram for explaining how addresses are allocated to RAMs in the same server.
- the replacement destination is FIG. 9 is a diagram for explaining a state in which an address is given by a combination of a time slot and an input symbol.
- FIGS. 109A and 109B are diagrams for explaining the storage capacity of the RAM in the same receiver.
- FIG. 109A shows the storage capacity of the RAM in a normal state
- FIG. 109B shows the pseudo storage capacity of the RAM in the case where the RAM operates as a partial write RAM.
- Fig. 110 is a diagram for explaining how data is written to and read from RAM at the same time in the receiver, using two banks of RAMs of three timeslots per node.
- FIG. 9 is a diagram for explaining an example of realizing a delay of one minute lead length for six minutes slot.
- FIG. 11 is a flowchart illustrating the timing of writing and reading data by the operation shown in FIG.
- FIG. 1 12 is a block diagram illustrating a configuration example of a convolutional encoder.
- FIG. 113 is a block diagram illustrating an example of a configuration of an encoding device, and is a block diagram illustrating a manner in which the order between input symbols for an interleaving bar is switched.
- FIG. 114 is a block diagram illustrating a simplified configuration of two adjacent element decoders constituting the decoding apparatus.
- FIG. 114 is a block diagram illustrating a configuration in which the connector has a symbol switching circuit.
- FIG. 115 is a block diagram illustrating a simplified configuration of two adjacent element decoders that constitute the decoding apparatus.
- FIG. 115 is a block diagram illustrating a configuration in which the soft-output decoding circuit has a symbol exchange circuit.
- FIG. BEST MODE FOR CARRYING OUT THE INVENTION hereinafter, specific embodiments to which the present invention is applied will be described in detail with reference to the drawings.
- digital information is encoded by an encoding device 1 provided in a transmitting device (not shown), and the output is input to a receiving device (not shown) via a memoryless communication path 2 with noise. Then, it is a data transmission / reception system applied to a communication model for decoding by the decoding device 3 provided in the receiving device.
- the coding apparatus 1 includes a parallel concatenated convolutional code (PCCC) using a trellis code such as a convolutional code as an element code, or a serial concatenated convolutional code (Seria lly).
- SCCC Concatenated Convolutional Codes
- TT CM multi-level modulation
- SCT CM serial Concatenated Trellis Coded Modulation
- the decoding device 3 decodes the code coded by the coding device 1, and is described in "Robertson, Villebrun and Hoeher," "A comparison of optim al and sub-optimal MAP decoding algorithms operating in the domain ”, IE EE Int. Conf. on Communications, pp. 1009-1013, June 1995”, the Max—Log—MAP algorithm or the Log—MAP algorithm (hereinafter the Max—Log—BCJR algorithm or The maximum posterior probability (Maximum A Posteriori probability; hereinafter, referred to as MAP) based on the LogB CJR algorithm is decoded, and the so-called probability, ⁇ , ⁇ , and soft output (soft-output) are decoded.
- a module that includes at least an array that rearranges the input data as a single element decoder, and is connected to multiple element decoders to perform iterative decoding.
- the decoding device 3 has a function of switching between a received value input to each element decoder and so-called extrinsic information as code likelihood, thereby providing an input for performing soft output decoding.
- Information to be performed is appropriately selected, and any code among PCCC, SCCC, TTCM or SCTCM can be decoded with the same configuration.
- each element decoder in the decoding device 3 is a Log-BC JR Al MAP decoding based on algorithm
- Encoding device and decoding device that perform encoding and decoding by 1PCCC
- an encoding device 1 ′ and a decoding device 3 ′ that perform encoding and decoding by PCCC shown in FIGS.
- a description will now be given of the encoding devices 1 and 2 and the decoding devices 3 and 3 for performing encoding and decoding by SCCC shown in FIGS.
- These encoding devices 1, 1, 1, 2 are positioned as examples of the encoding device 1, and the decoding devices 3, 3 ′ ′ are positioned as an example of the decoding device 3.
- the decoding devices 3 ′ and 3 ′ ′ can be configured by connecting element decoders.
- an encoding device 1 ′ that performs encoding by PCCC, and this encoding device 1, A decoding device 3 ′ that decodes a code according to.
- the encoder 1 ′ includes a delay unit 11 for delaying input data, two convolution encoders 12 and 14 for performing convolution operation, and Some have an interleaver 13 that rearranges the evening order.
- the coding device 1 performs a parallel convolution operation with a coding rate of "1/3" on the input 1-bit input data D1 and outputs a 3-bit output data.
- D4, D5, and D6 are generated, for example, a binary phase shift keying (hereinafter, referred to as BPSK) modulation method and a quadrature phase shift keying (hereinafter, referred to as QPSK).
- BPSK binary phase shift keying
- QPSK quadrature phase shift keying
- the signal is output to the outside via a modulator (not shown) that performs modulation by a modulation method.
- the delay unit 11 is provided for adjusting the timing at which the 3-bit output data D 4, D 5, and D 6 are output.
- the delay unit 11 outputs the 1-bit input data D 1.
- the input data D 1 is delayed by the same time as the processing time required by the interleaver 13.
- the delay unit 11 outputs the delayed data D2 obtained by delaying to the outside as an output data D4, and supplies the delayed data D2 to the convolutional encoder 12 in the subsequent stage.
- the convolutional encoder 12 Upon receiving the 1-bit delay data D 2 output from the delay unit 11, the convolutional encoder 12 performs a convolution operation on the delay data D 2 and outputs the operation result to the output data D 5. And output to the outside.
- the input receiver 13 receives input data D 1 composed of one bit sequence, rearranges the order of each bit constituting the input data D 1, and generates the generated input data D 1. —The data D 3 is supplied to the subsequent convolutional encoder 14.
- the convolutional encoder 14 When the convolutional encoder 14 receives the 1-bit interleave data D3 supplied from the interleaver 13, the convolutional encoder 14 performs a convolution operation on the interleave data D3. And outputs the operation result to the outside as output data D6.
- an encoding device 1 ′ When such an encoding device 1 ′ receives 1-bit input data D 1, it receives the input data D 1 as output data D 4 of the tissue component and outputs it as it is via the delay unit 11. And the output data D5 obtained as a result of the convolution operation of the delayed data D2 by the convolutional encoder 12 and the convolution operation of the interleaved data D3 by the convolutional encoder 14. By outputting the resulting output data D 6 to the outside, the concatenated convolution operation with a code rate of “1/3” as a whole is performed. Perform the calculation.
- the data encoded by the encoding device 1 ′ is mapped to signal points by a modulator (not shown) based on a predetermined modulation method, and is output to the receiving device via the memoryless communication channel 2.
- a decoding device 3 ′ that decodes a code by the encoding device 1 ′ includes two soft output decoding circuits 15 and 17 that perform soft output decoding and an input data That has an interleaver 16 that rearranges the order of input data, two dinterleavers 18 and 20 that restore the order of input data, and an adder 19 that adds two data. There is.
- the decoding device 3 ′ estimates input data D 1 in the encoding device 1 from the received value D 7 which is considered as a soft input due to the influence of noise generated on the memoryless communication channel 2.
- the decrypted data is output as D13.
- the soft output decoding circuit 15 is provided corresponding to the convolutional encoder 12 in the encoding device 1 ', and performs MAP decoding based on Log-BCRC.
- the soft output decoding circuit 15 receives the soft input received value D 7 and generates a priori probabilistic information (a priori proba bili li) for the soft input information bit output from the dinosaur filter 18. ty information) D8 is input, and soft output decoding is performed using these received values D7 and prior probability information D8. Then, the soft-output decoding circuit 15 generates external information D 9 for the information bit determined by the code constraint condition, and outputs the external information D 9 to the subsequent stage in-line receiver 16 as a soft output.
- the interleaver 16 uses the same permutation position information as the interleaver 13 in the encoder 1 ′ for the external information D 9 for the soft input information bits output from the soft output decoding circuit 15. I give you one night leave.
- the interleaver 16 outputs the data obtained by the interleaving as prior probability information D 10 for the information bit in the subsequent soft output decoding circuit 1 ⁇ , and outputs the data to the subsequent adder 19. .
- the soft output decoding circuit 17 is provided corresponding to the convolutional encoder 14 in the encoder 1 ′, and, like the soft output decoding circuit 15, is a MAP decoding based on the Log-BCJR algorithm. I do.
- the soft-output decoding circuit 17 receives the soft-input received value D 7 and the prior probability information D 10 for the soft-input information bits output from the interleaver 16, and receives these received values D 7. 7 and prior probability information D Soft output decoding is performed by using 1 and 0. Then, the soft-output decoding circuit 17 generates external information D 11 for the information bit obtained by the constraint condition of the code, and outputs the external information D 11 to the interleaver 18 as a soft output. , And output to the adder 19.
- Dinverter 18 restores the bit array of interleave data D 3 interleaved by interleaver 13 in encoder 1 ′ to the original bit array of input data D 1
- the external information D 11 of the soft input output from the soft output decoding circuit 17 is subjected to a din-one-leaving.
- Dinter leaver 18 outputs the data obtained by deinterleaving as prior probability information D 8 for the information bit in soft output decoding circuit 15.
- the adder 19 includes a priori probability information D 10 for the soft input information bit output from the receiver 16 and an external probability information D 10 for the information bit output from the soft output decoding circuit 17.
- the information D 1 1 is added.
- the adder 19 outputs the obtained data D 12 as a soft output to the subsequent interleaver 20.
- the din / leaver 20 returns the bit array of the interleave data D3 interleaved by the interleaver 13 in the encoders 1 and 2 to the original bit array of the input data D1. Then, the soft output data D 12 output from the adder 19 is subjected to a dive reeve.
- the data leaver 20 outputs the data obtained by the data leave to the outside as decoded data D 13.
- Such a decoding device 3 ′ is provided with soft output decoding circuits 15, 17 corresponding to those of the convolutional encoders 12, 14 in the encoding devices 1, 2, thereby providing a high decoding complexity code. Can be decomposed into elements having low complexity, and the characteristics can be sequentially improved by the interaction between the soft output decoding circuits 15 and 17.
- the decoding device 3 Upon receiving the received value D7, the decoding device 3 'performs iterative decoding with a predetermined number of repetitions, and outputs decoded data D13 based on soft-output external information obtained as a result of this decoding operation. You.
- an encoding device that performs encoding by TTCM includes, for example, an 8-phase shift keying (hereinafter, referred to as 8 PSK) modulation at the last stage of the encoding device 1 ′. It can be realized by providing a vessel. Also, TT The decoding device that decodes the code by the CM can be realized by the same configuration as the decoding device 3 ′, and directly receives the in-phase component and the quadrature component symbol as reception values.
- 8 PSK 8-phase shift keying
- the encoder 1 ′ ′ includes a convolutional encoder 31 for encoding a code called an outer code, an interleaver 32 for rearranging the order of input data, and an inner code And a convolutional encoder 33 that performs encoding of a code called “convolutional code”.
- the encoding apparatus 1 ′ performs a cascade convolution operation with an encoding rate of “1” on the input 1-bit input data D 21, and outputs 3-bit output data.
- D 26, D 27, and D 28 are generated and output to the outside via a modulator (not shown) that performs modulation by, for example, the BP SK modulation method or the QP SK modulation method.
- convolutional encoder 31 Upon input of 1-bit input data D 21, convolutional encoder 31 performs a convolution operation on this input data D 21, and outputs the operation result as 2-bit encoded data D 2 2 , D 23 to the subsequent interleaver 32. That is, the convolutional encoder 31 performs convolutional operation with an encoding rate of “1/2” as outer code encoding, and generates the encoded data D 22 and D 23 at the subsequent interleaver. 3 to 2
- the interleaver 32 inputs the coded data D 22 and D 23 composed of two bit sequences supplied from the convolutional coder 31, and these coded data D 22: D
- the order of the bits constituting 23 is rearranged, and the generated interleaved data D 24 and D 25 composed of the two bit sequences are supplied to the convolutional encoder 33 at the subsequent stage.
- the convolutional encoder 33 When the convolutional encoder 33 receives the 2-bit in-leave data D 24 and D 25 supplied from the in-leave receiver 32, the convolutional encoder 33 outputs the interleave data D 24 and D 25 to the interleave data D 24 and D 25.
- the convolution operation is performed, and the operation result is output to the outside as a 3-bit output data D26, D27, and D28. That is, the convolutional encoder 33 performs a convolution operation with a coding rate of “2/3” as the encoding of the inner code, and outputs The data D 26, D 27, and D 28 are output to the outside.
- Such an encoding device 1 ′ ′ performs a convolution operation with an encoding rate of “1/2” as the encoding of the outer code by the convolutional encoder 31, and the convolutional encoder 33 encodes the inner code.
- the data encoded by the encoding device 1 ′′ is subjected to signal point matching based on a predetermined modulation method by a modulator (not shown), and is output to the receiving device via the memoryless communication channel 2.
- a decoding device 3 ′ ′ that decodes a code by the encoding device 1 ′ ′ has two soft output decoding circuits 34 and 36 that perform soft output decoding, and Some include a sorter for reversing the order 35 and an interleaver 37 for reordering the input data.
- the decoding device 3 ′ estimates the input data D 21 in the encoding device 1 ′ ′′ from the received value D 29 which is a soft input due to the influence of noise generated on the memoryless communication channel 2, and outputs the decoded data D 3 Output as 6.
- the soft output decoding circuit 34 is provided corresponding to the convolutional encoder 33 in the encoding device 1 ′ ′, and performs MAP decoding based on L 0 g—B C J R.
- the soft output decoding circuit 34 receives the soft input reception value D 29, and inputs prior probability information D 30 for the soft input information bit output from the interleaver 37, and receives these reception values D 29.
- MAP decoding based on the Log-BC JR algorithm is performed, and soft output decoding of the inner code is performed.
- the soft output decoding circuit 34 generates external information D31 for the information bits obtained under the code constraint conditions, and outputs this external information D31 to the subsequent din receiver 35 as a soft output. I do.
- the external information D31 corresponds to the interleave data D24 and D25 which are interleaved by the interleaver 32 in the encoding device 1 '.
- the DIN receiver R 35 is a bit array of D 24 and D 25 interleaved by the receiver R 32 in the encoders 1 and 2, respectively.
- the soft output is returned to return to the bit array of the original encoded data D22 and D23.
- the interleaving of the soft input external information D 31 output from the decoding circuit 34 is performed.
- the soft output decoding circuit 36 is provided corresponding to the convolutional encoder 31 in the encoding device 1 ′, and performs MAP decoding based on Log-B CJR.
- the soft output decoding circuit 36 inputs the prior probability information D 32 for the sign bit of the soft input output from the interleaver 35 and outputs the prior probability information D 33 for the information bit having a value of “0”.
- the soft-output decoding circuit 36 uses the prior probability information D32 and D33, MAP decoding based on the L0g-BCJR algorithm is performed, and soft output decoding of the outer code is performed.
- the soft-output decoding circuit 36 generates external information D34 and D35 obtained by the code constraint condition, outputs the external information D34 to the outside as decoded data D36, and outputs the external information D35 to the interface. Output as soft output to Lever 37.
- the interleaver 37 has the same replacement position as the interleaver 32 in the encoder 1 ′ ′ with respect to the external information D 35 for the sign bit as the soft input output from the soft output decoding circuit 36. Perform an information-based interleave.
- the interleaver 37 outputs the data obtained by the in-leave and re-leave as prior probability information D30 for the information bits in the soft-output decoding circuit 34.
- Such a decoding device 3 ′ ′ is provided with soft output decoding circuits 36, 34 corresponding to those of the convolutional coder 31, 33 in the encoding device 1 ′ ′, so that it is similar to the decoding device 3 ′.
- the code having high decoding complexity can be decomposed into elements having low complexity, and the characteristics can be sequentially improved by the interaction between the soft output decoding circuits 34 and 36.
- the decoding device 3 ′ ′ Upon receiving the reception value D 29, the decoding device 3 ′ ′ performs iterative decoding with a predetermined number of repetitions, and decodes the decoded data D 36 based on the soft-output external information obtained as a result of this decoding operation. Is output.
- an encoding device that performs encoding by SCT CM can be realized by including a modulator that performs modulation by, for example, the 8PSK modulation method at the last stage of the encoding device 1 ′′ ′.
- a decoding device that decodes a code using SCT CM can be realized with the same configuration as the decoding device 3 ′ ′. The symbol of the orthogonal component is directly input.
- the decoding device 3 shown as an embodiment of the present invention includes a module including at least a soft-output decoding circuit and an interleaver or a din-leaver as shown by a broken line in FIG. 7 or a broken line in FIG. It decodes an arbitrary code among PCCC, SCCC, TTCM or SCTCM by connecting multiple element decoders.
- the deinterleaver since the deinterleaver reorders data based on the replacement position information opposite to the interleaver, it can be imitated as a form of the interleaver.
- any element decoder may be used as long as it has a soft-output decoding circuit and an interleaver, and performs an interleave process and an interleave process, and switches the function between the interleaver and the interleaver. It can be realized by doing. Therefore, in the following, it is assumed that the interleaver has the function of the interleaver unless otherwise required.
- M states transition states representing the contents of the shift registers provided in each element encoder in the encoding device 1 are represented by m (0, 1,%) As necessary.
- the element decoder 50 schematically shown in FIG. 10 is configured as one chip by integrating each part on a single semiconductor substrate as a large-scale integrated circuit (hereinafter, referred to as LSI).
- the element decoder 50 includes a control circuit 60 for controlling each unit, a decoded reception value selection circuit 70 for selecting a reception value to be decoded, an edge detection circuit 80 for detecting the head of a frame, and a soft output for performing soft output decoding.
- FIG. 11 shows the details of the left half of the element decoder 50 shown in FIG. 11
- FIG. 12 shows the details of the right half.
- Control circuit 60 decodes the received value selection circuit 70, the soft-output decoding circuit 90, Intari - Bruno 1 00, the address storage circuit 1 1 0 and nine selector 1 20 2, 1 2 0 3, 1 2 04 , 1 2 0 5 , 1 2 0 1 2 0 ⁇ , 1 2 0 8) 1 2 0 9 , 1 20!. It generates and supplies various kinds of information, receives information from the address storage circuit 110, and controls the operation of each unit.
- control circuit 60 selects a reception value for causing the decoding reception value selection circuit 70 to select a decoding reception value TSR which is a reception value to be decoded from the reception values R (reception values TR). Generate and supply information CRS.
- control circuit 60 determines whether the data input as the received value R is actually a received value or external information to the soft output decoding circuit 90, and furthermore, the encoding device 1 Received value format information that indicates the format of the received value R, such as whether it is an I / Q value in the case of encoding using CM or SCT CM CRT Y And the prior probability information format information CAP P indicating the format of the prior probability information such as whether the prior probability information is input in units of bits or symbols, and the code of the element encoder in the encoding device 1.
- Coding rate information CRAT indicating the coding rate
- generator matrix information CG indicating the generating matrix of the elementary encoder in the coding apparatus 1
- coding apparatus 1 performing coding by TTCM or SCT CM.
- signal point arrangement information CSIG indicating the arrangement of signal points is generated and supplied.
- control circuit 60 includes, for the interleaver 100, an interleaver type information CINT indicating what type of interleave is to be performed, and an interleaver indicating the interleaver length.
- puncture cycle information CNEL indicating a puncture cycle and puncture pattern information CNEP indicating a puncture pattern are generated and supplied. Further, the control circuit 60 generates and supplies operation mode information CBF indicating an operation mode described later to the interleaver 100.
- the control circuit 60 interleaves the address storage circuit 110 with the interleaver type.
- control circuit 60 to the six selectors 1 20 2, 1 2 03, 1 204, 1 20 5j 1 20 a, 1 20 7, and supplies the operation mode information CB F, 3 single selector 1 20 8 , 120 9, 120 i.
- verification mode information CTHR indicating whether or not a verification mode is described later is supplied.
- control circuit 60 inputs the read address data ADA which is the address data of the replacement destination which is referred to by the interleaver 100 held in the address memory circuit 110.
- the operation of each unit is controlled, and the control and operation of writing address data to the address storage circuit 110 are performed.
- the decoded reception value selection circuit 70 is provided for decoding an arbitrary code, as described later, and is input based on reception value selection information CRS supplied from the control circuit 60. From the received values TR, select the decoded received value TSR. The decoded received value selection circuit 70 supplies the selected decoded received value TSR to the soft output decoding circuit 90.
- the decoded received value selection circuit 70 is configured such that the received value TR is composed of six systems of the received values TRO, TR1, TR2, TR3, TR4, and TR5. Are selected as decoded received values TSR 0, TSR 1, TSR 2, and TSR 3. For example, as shown in FIG. 13, assuming that there are four selectors 71, 72, 73, and 74. Can be realized. At this time, the received value selection information CRS supplied from the control circuit 60 is individually given to each of the selectors 71, 72, 73, and 74, and the received value selection information CR S0, CRS 1, CRS 2, and CRS 3.
- the selector 71 selects a predetermined reception value from the reception values TR 0, TR 1, TR 2, TR 3, TR 4, and TR 5 based on the reception value selection information CRS 0, and performs decoding and reception.
- the value is supplied to the soft output decoding circuit 90 as the value TSR0.
- the selector 72 selects a predetermined reception value from the reception values TR 0, TR 1, TR 2, TR 3, TR 4, and TR 5 based on the reception value selection information CRS 1, and decodes The value is supplied to the soft output decoding circuit 90 as the value TSR 1.
- the selector 73 selects a predetermined reception value from the reception values TR 0, TR 1, TR 2, TR 3, TR 4, and TR 5 based on the reception value selection information CRS 2, and decodes the reception value. It is supplied to the soft output decoding circuit 90 as TSR 2.
- the selector 74 selects a predetermined reception value from the reception values TR 0, TR 1, TR 2, TR 3, TR 4, and TR 5 based on the reception value selection information CRS 3, and decodes the selected value.
- the received value is supplied to the soft output decoding circuit 90 as the received value TSR3.
- the decoded received value selection circuit 70 selects the decoded received value TSR based on the received value selection information CRS supplied from the control circuit 60, and supplies it to the soft output decoding circuit 90.
- the edge detection circuit 80 receives an externally supplied interleave start position, that is, an interleave start position signal ILS (interleave start position signal TILS) indicating the beginning of a frame. Then, the head of the frame constituting the input received value TR is detected. Edge detection circuit 80 supplies a Etsu di signal TE ILS indicating the beginning of the detected frame to the soft-output decoding circuit 9 0 and the selector 1 20 5.
- ILS interleave start position signal
- the edge detection circuit 80 can be realized as having, for example, a register 81 and an AND gate 82 as shown in FIG.
- the register 81 holds, for example, one bit of the interleave start position signal TILS for one clock.
- the register 81 supplies the held delay interleaving start position signal TILSD to the AND gate 82.
- the AND gate 82 outputs the interleave start position signal TILS and the inverted interleave start position signal TIL SD, which is the interleave start position signal TILS that is supplied from the register 81 and is the one-clock preceding interleave start position signal TILS. Perform a logical conjunction.
- a ND gate one DOO 82 supplies the obtained logical product to the soft-output decoding circuits 9 0 and the selector 1 2 0 5 as edge signal TEILS.
- the edge detection circuit 80 only needs to detect that the interleaving start position signal TILS supplied from the outside is switched from “0” to “1”, and by taking the logical product by the AND gate 82, It is possible to detect that the beginning of the frame constituting the received value TR has been input.
- the soft-output decoding circuit 90 outputs the decoded received value TSR supplied from the decoded received value selection circuit 70 and external information or interleaved data supplied from outside as prior probability information. Perform MAP decoding based on the Log-BCJR algorithm using the leave data T EXT).
- the soft-output decoding circuit 90 generates the received value format information CRTY supplied from the control circuit 60, the prior probability information format information C APP, the coding rate information CRAT, In addition to the matrix information CG and, if necessary, the signal point arrangement information CSIG, erasure information ERS (erasure information TER S) indicating the puncture pattern supplied from the outside, and prior probability information erasure information EAP (erasure of prior probability information)
- the decoding process is performed using the information TEAP), the termination time information TNP (termination time information TTNP) indicating the termination time of the code, and the termination state information TNS (termination state information TTNS) indicating the termination state.
- the soft output decoding circuit 90 supplies the soft output SOL and the external information SOE obtained as a result of the decoding process to the selector 120. At this time, the soft output decoding circuit 90 outputs the information for the information symbol or the information bit and the code symbol or the code bit based on the output data selection control signal I TM (output data selection control signal CI TM) supplied from the outside. And selectively output the information. Also, when the soft output decoding circuit 90 makes a hard decision, the decoded value hard decision information SDH obtained by performing a hard decision on the soft output which is the decoded value and the received value are obtained by performing a hard decision. Outputs the received value hard decision information S RH to the outside. Also at this time, the soft output decoding circuit 90 selectively outputs the information for the information symbol or the information bit and the information for the code symbol or the code bit based on the output data selection control signal CITM.
- the soft output decoding circuit 90 can also delay the reception value TR, external information or interleaved data TEXT, and the edge signal TE ILS supplied from the edge detection circuit 80, respectively. it can.
- the soft-output decoding circuit 90 supplies the delayed received value SDR obtained by delaying the received value TR to the selector 1 20 3, 1 2 0 6, delay the external information obtained by delaying the external information or fin evening over Reeve data TEXT supplying SDEX to the selector 1 20 2 supplies the delay edge signal SD ILS obtained by delaying the edge signal TEILS to the selector 1 20 5.
- interleaver 1 00 In the evening one interleaver 1 00 is to de Isseki TII supplied from the selector 1 2 0 4, interface one leave that based on fin evening the same substitution position information and one interleaver in the encoder 1 (not shown) Or Dinterleaving is performed to return the bit array of the interleaved data interleaved by the interleaver in the encoder 1 to the bit array of the original data.
- Inn Yuichi Lever 100 It functions as an interleaver or an interleaver based on the interleave mode signal DIN (interleave mode signal CD IN) supplied from the outside.
- Inta one Lever 1 00 inputs the fin evening one leave the start position signal TIS supplied from the selector 1 2 0 5, the address storage circuit 1 1 0, Adoresude - the evening Adoresu giving I AA
- the address data stored in the address storage circuit 110 is read as read address data ADA, and interleaving or dinterleaving is performed based on the read address data ADA.
- the interleaver 100 uses the interleaver type information CINT supplied from the control circuit 60, the interleaver length information CINL, and the interleaver input / output replacement information CIPT. And interleave or interleave.
- Inta one Lever 1 00 supplies the fin evening one interleaver output data 1 10 obtained fin evening Ribu or di interleave selector 1 20 7.
- inter one interleaver 1 00 may be either one of the data TD I of the received value TR or delayed received value SDR supplied from the selector 1 20 3 slow cast.
- the input / output unit 100 delays the data TDI based on the operation mode information CBF supplied from the control circuit 60.
- Tariba 1 00 supplies Inta one sleeve length delayed received value I DO obtained by delaying the de Isseki TD I to the selector 1 20 6.
- the in-box receiver 100 includes termination position information CNF T, termination period information CNF L, termination state information CN FD, and puncture cycle information CNE L supplied from the control circuit 60.
- the puncture pattern information CNEP when a plurality of the element decoders are connected, the termination time information IGT and the termination state indicating the termination time and the termination state of the code in the next element decoder. It generates information IGS, erasure position information IGE indicating a puncture position of the code, and interposer non-output position information INO.
- the interleaver 100 has the generated termination time information IGT, termination state information IGS, and erase position information I.
- the GE, interleaver non-output position information IN ⁇ , and the delay interleave start position signal IDS are synchronized with the beginning of the frame as generation next-stage information, and the selector 120i. To supply.
- the address storage circuit 110 includes, for example, a plurality of banks of RAM (Random Access Memory) and a selection circuit, and is referred to when interleaving or interleaving by the interleaver 100 is performed.
- the replacement position information of the data to be stored is stored as address data.
- the address data stored in the address storage circuit 110 is read by the interleaver 100 when the address of the address storage circuit 110 is designated as the address data IAA by the interleaver 100. Read as ADA.
- Writing of address data to the address storage circuit 110 is performed by the control circuit 60, and the address data is written by specifying the address of the address storage circuit 110 as the address CIAD. Written as data C IWD.
- the address storage circuit 110 may be provided inside the interface 100.
- the element decoder 50 performs the in-leave-leaving process or the dinter-leave process by using both the in-time one-reaver 100 and the address storage circuit 110.
- the selector 120 selects one of the soft output S 0 L supplied from the soft output decoding circuit 90 and the external information S 0 E based on the output data selection control signal CITM, and outputs the data T to the selector 1 20 2 as LX. That is, the selector 120 t determines whether the soft-output decoding circuit 90 should output external information in the process of iterative decoding or should output a soft output as a final result. It is provided in order to perform.
- the selector 1 20 based on the operation mode information CBF, and delay the external information SDEX the soft-output decoding circuit 9 0 to be supplied, the selector 1 2 0! Of the data T LX supplied from either one select, as the data TD LX, to the selector 1 20 4, 1 20 7.
- the element decoder 50 has, for example, six operation modes. First, the soft output decoding circuit 90 and the input / output leaver 100 perform normal soft output decoding processing and input / output leave processing, respectively. The second is a mode in which only the soft output decoding circuit 90 performs normal soft output decoding processing. Third, only the interleaver 100 performs normal interleave processing.
- the soft output decoding circuit 90 and the interleaver 100 do not perform ordinary soft output decoding processing and interleaving processing, but function as delay circuits.
- Fifth there is a mode in which only the soft output decoding circuit 90 functions as a delay circuit without performing ordinary soft output decoding processing.
- Sixth there is a mode in which only the interleaver 100 functions as a delay circuit without performing normal interleave processing.
- These operation modes are determined by the control circuit 60 and are supplied to each section as operation mode information CBF.
- the first to third modes are collectively referred to as a normal mode
- the fourth to sixth modes are collectively referred to as a delay mode, as necessary.
- the selector 1 2 0 2 the operation mode information CBF is, the processing time and the same time soft-output decoding circuit 9 0 required delay, fin evening one interleaver 1 0 0 it takes processing time and the same time If the signal indicates a delay or a delay mode in which either the processing time required by the soft-output decoding circuit 90 and the interleaver 100 0 or the delay between simultaneous processing should be performed, the delay external information SDEX is selected.
- the operation mode information CBF is not delayed by the soft-output decoding circuit 90 and / or the input / output switch 100 0, and the soft-output decoding circuit 90 and / or the interleaver 100 0 is not performed.
- the data TLX is selected and output. That is, the selector 1 2 0 2, or the operating mode of the element decoder 5 0 that is the delay mode, or, which whether the normal mode, is provided for determining, depending on the respective operating mode , Select the output to be output.
- Selector 1 2 0 3 based on the operating mode one de information CBF, the received value TR, of the delayed received value SDR supplied from the soft-output decoding circuit 9 0, selects one, de Isseki The TDI is supplied to the in-box receiver 100. Specifically, the selector 1 2 0 3, the operation mode information CBF performs only processing by the interlink one interleaver 1 0 0 In the case of the normal mode or the delay mode in which the interleaver 100 should delay the same time as the required processing time, the received value TR is selected and output, and the operation mode is set. If the information CBF indicates any other normal mode or delay mode, select and output the delayed reception value SDR.
- the selector 1 2 0 3 as data to be input to the fin evening one interleaver 1 0 0, the soft-output decoding circuit 9 soft-output decoding or soft-output decoding circuit 9 0 it takes processing time and the same time by 0 This is provided to determine whether or not to use the data after the delay, and selects the data to be output according to each operation mode.
- Selector 1 2 0 4 based on the operation mode information CBF, Chi sac and external information or fin evening one Ribude Isseki TEXT, the data TDLX supplied from the selector 1 2 0 2, selects either The data is supplied to the interleaver 100 as data TII.
- the selector 1 2 0 4 the operation mode information CBF is, the normal mode performs only processing that the fin evening one interleaver 1 0 0 or, the processing time fin evening one interleaver 1 0 0 it takes If it indicates a delay mode in which the same time delay should be performed, external information or interleaved text is selected and output, and the operation mode information CBF is set to the other normal mode or delay mode.
- select and output the data TDLX If it indicates, select and output the data TDLX. That is, the selector 122 4 performs the same process as the soft output decoding process by the soft output decoding circuit 90 or the processing time required by the soft output decoding circuit 90 as the data to be input to the interleave and '100'. This is provided to determine whether or not to use the data after the delay, and selects the data to be output according to each operation mode.
- the edge signal TEILS supplied edge detection circuit 8 0 to, among the delayed edge signal SDILS supplied from the soft-output decoding circuit 9 0, or One is selected and supplied to the interleaver 100 as the interleave start position signal TIS.
- the selector 1 2 0 5 the operation mode information CBF is, normal mode or inter one interleaver 1 0 0 it takes processing time and the same time performs only the process by in the evening some truth ICHIBA 1 0 0 If it indicates the delay mode in which the delay should be performed, the edge signal TEILS is selected and output, and the operation mode information CBF indicates the other normal mode or delay mode.
- the selector 1 2 0 5 in the evening as data input to one Reno 1 0 0, and during the processing time of the soft output decoding circuit 9 soft-output decoding or soft-output decoding circuit 9 0 by 0 takes This is provided to determine whether or not to use the one delayed by the same time.
- the output data is selected according to each operation mode.
- the selector 120 ⁇ selects the delayed reception value SDR supplied from the soft output decoding circuit 90 and the interleave long delay reception value IDO supplied from the interleaver 100. of the, selects one, and supplies to the selector 1 2 0 8 as the delayed received value TDR.
- the selector 1 2 0 6, the operation mode - de information CBF is, the normal mode performs only processing by the soft-output decoding circuit 9 0, or, the soft-output decoding circuit 9 0 it takes processing time and the same time If it indicates the delay mode in which the delay should be performed, select and output the delayed reception value SDR, and if it indicates any other normal mode or delay mode, Selects and outputs the interleave length delay reception value IDO.
- selectors 1 2 0 6 as to be output data, used after subjected to interleaver 1 0 0 Inta one leave processing or in the evening one interleaver 1 0 0 processing time and the time required delay by It is provided to determine whether or not to output data according to each operation mode.
- Selector 1 2 0 7 based on the operation mode information CBF, the fin evening one interleaver output data II 0 supplied from Inta one interleaver 1 0 0, among the data TDLX supplied from the selector 1 2 0 2 , selects one, and supplies to the selector 1 2 0 9 as soft output TSO.
- the selector 1 2 0 7, the operation mode information CBF is, the normal mode performs only processing by the soft-output decoding circuit 9 0, or, the processing time of the same time delay the soft-output decoding circuit 9 0 it takes If it indicates the delay mode to be performed, select and output TDLX overnight.If it indicates any other normal mode or delay mode, output the interleaver. Selects and outputs II 0.
- the selector 1 2 0 7, as data to be output, those performed in the evening one interleaver 1 0 0 fin evening one leave processing or Intariba 1 0 0 is processing time and the time required delay by This is provided to determine whether or not the data is to be output. The output data is selected according to each operation mode.
- the selector 1 20 based on the verification mode information C THR, the delayed received value TDR supplied selector 1 2 0 6 or et of a through signal that is transmitted by the signal line 1 30, either one Is selected and output to the outside as the delayed reception value TRN.
- the delayed received value TRN is output as the delayed received value RN. That is, selector 1 2 0 8 and is provided either to output the delayed received value for the next stage element decoder, whether to validate the system, in order determined.
- the selector 1 20 9, based on the verification mode information CT HR, the soft output TS 0 which is either found supply selector 1 20 Ma, of the Solo signal transmitted by the signal line 1 30, either Select and output to the outside as soft output TINT.
- the soft output TINT is output as soft output INT. That is, the selector 1 20 9 and is provided either to output the soft output for the next element decoder, a row Unoka verification system for determining.
- the erase position information IGE and the interleaver non-output position information are provided. Either INO and the next stage information consisting of the delay in-leave start position signal IDS or the through signal transmitted through the signal line 130 are selected, and the next stage end time information is selected. It is output to the outside as T TNPN and next-stage termination state information TTNSN, next-stage erasure position information TERSN and next-stage prior probability information erasure information TE APN, and next-stage interleave start position signal TILSN.
- next-stage termination time information TTNPN, next-stage termination state information TTNSN, next-stage erasure position information TERSN, next-stage prior probability information erasure information TEAPN, and next-stage interleave start position signal TILSN are: Next-stage termination time information TNPN, next-stage termination state information TNSN, next-stage erasure location information ERSN, next-stage prior probability information erasure information EAPN, and next-stage start-up location Output as signal ILSN. That is, selector 1 2 0! . Is provided to determine whether to output the next-stage information for the next-stage element decoder or to verify the system.
- the signal line 130 is mainly used when a decoding device 3 similar to the above-described decoding devices 3 ′, 3, and ′ is configured by connecting a plurality of element decoders 50. It is used for verifying the system.
- the signal line 130 has a reception value TR, external information or in-leaving data TEXT, erasure information TER S, prior probability information erasure information TE AP, termination time information TTNP, termination state information T TN S, and interleaving. configured it it by bundling signal lines for transmitting the start position signal TILS, these signal selector 1 20 8, 1 2 0 s , 1 20 i. Supply to
- Such an element decoder 50 includes, for example, a module including at least a soft-output decoding circuit and at least an interleaver or an interleaver as shown by a broken line in FIG. 7 or a broken line in FIG. Are equivalent.
- a decoding device 3 capable of decoding an arbitrary code among PCCC, SCCC, TTCM or SCTCM can be configured.
- Various features relating to the entire element decoder 50 will be further described in "4."
- the soft output decoding circuit 90 includes a code information generation circuit 15 1 that generates code information of an element encoder in the encoding device 1, and a puncture pattern in the encoding device 1.
- the internal erasure information generation circuit 152 for generating internal erasure information as shown in the figure
- the termination information generation circuit 153 for generating the termination information in the encoder 1, and the received value to be input for the decoding process.
- a priori probability information is selected, a received value and a priori probability information selection circuit 154 for replacing a position where no code output exists with a symbol with a likelihood of “0”, and the received data and delay data
- a storage circuit for receiving and storing the received data and the delay 155, an Ia calculation circuit 156 for calculating a log likelihood Ia as a first log likelihood, and a calculation circuit 1 for calculating An Ia distribution circuit 157 that distributes the log likelihood Ia, and a second An I line calculation circuit 158 for calculating a log likelihood I, which is a number likelihood, an I? Calculation circuit 159 for calculating a log likelihood I ?, a third log likelihood, and a calculated log likelihood I?
- Storage circuit 160 that stores the degree I, and a soft output calculation circuit that calculates the log soft output I person Circuit 161, a received value or prior probability information separating circuit 162 for separating the received value and the prior probability information, an external information calculating circuit 163 for calculating the external information, and a log soft output I input It has an amplitude adjusting and clipping circuit 164 for adjusting the amplitude and clipping to a predetermined dynamic range, and a hard decision circuit 165 for hard-deciding a soft output and a received value as decoded values.
- FIG. 16 shows details of the left half portion of the soft output decoding circuit 90 shown in FIG. 16, and FIG. 1 ⁇ shows details of the right half portion.
- the code information generating circuit 151 generates code information of the elementary encoder in the encoding device 1 based on the coding rate information CR AT and the generated matrix information CG supplied from the control circuit 60. Specifically, the code information generation circuit 151 includes input bit number information IN indicating the number of input bits of the element encoder in the encoding device 1 and convolution of the element encoder in the encoding device 1.
- the type information WM indicating whether the convolutional encoder is of the so-called Wozencraft type or Massey type
- the elementary encoder of the encoding device 1 In a shift register, that is, memory number information MN indicating the number of memories representing states (transition states), and a trellis which is a state transition diagram of an element encoder in the encoding device 1, an input along a time axis for each branch.
- the branch input / output information BI0 indicating output information and the output from the element encoder in the encoding device 1 are present, and the validity of the output position indicating the existence of the corresponding received value is present. It generates the output position information P E.
- the Posencraft-type convolutional encoder is composed of a delay element and a combinational circuit, and the delay element stores the data in time series.
- Examples of Bozenkura shift type convolutional encoder for example, as shown in FIG. 1 8, 4 and Tsunoshi full Torejisuta 2 0 1 t, 2 0 12 , 2 0 1 2 0 1 4, 1 6 pieces of exclusive Logical OR circuit 2 0 2 2 0 2 2 , 2 0 2 2 0 2 4 , 2 0 2 5 , 2 0 2 2 0 2 7 , 2 0 2 8 , 2 0 2 9 , 2 0 210, 2 0 211, 2 0 2, 2 0 2 i 3, 2 0 2 M, 2 0 2, 2 0 2 16 and 2 0 aND gates G 0 [0], GB [ 0], GB [1], GB
- the generator matrix G of this convolutional encoder is expressed by the following equation (27). In the following equation (27), GB (D), G1 (D), G2
- G3 (D) G 3 [ ⁇ ] + G 3 [l] D + G 3 [2] D 2 + G 3 [s] D 3 + G3 [4] D 4 (31)
- examples of the convolutional encoder Po Zen craft type for example, as shown in FIG. 1 9, three shift registers 2 03!, 203 2, and 203 3, 1 two exclusive-OR circuits 2 04 2042, 204 a, 204 4 , 204 5 , 204 ⁇ , 204 7, 204 8 , 2049, 20410, 20411, 204 i 2 and 15 AND gates G 1 [0], G 1 [1 ], G 1 [2], G 1 [3], G 1 [4], G 2 [0], G 2 [1], G 2 [2], G 2 [3], G 2 [4], A combinational circuit represented by G 3 [0], G 3 [1], G 3 [2], G 3 [3], G 3 [4], and a code rate of “2/3” Some perform convolution operations.
- G il (D), G 21 (D), G 31 (D), G 12 (D), G 22 (D), and G 32 (D) are It is expressed by the following equations (33) to (38).
- G (32) G 12 (D) G 22 (D) G 32 (D)
- GII (D) GI [O] + GI [I] D + GI [2] D 2
- the Massey-type convolutional encoder is composed of a delay element and a combinational circuit, and is configured to output one of the input bits as a tissue component as it is. That is, data is not stored in the delay element in time series.
- Examples of Matsushi I type convolutional encoder for example, as shown in FIG. 2 0, and evening three shift Torejisu 20 5 ,, 20 5, 2 0 5 3, four exclusive OR circuits 2 0 6 !
- the combinational circuit changes, the code configuration changes, and the maximum number of states is "2 3 8".
- the generator matrix G of this convolutional encoder is expressed by the following equation (39). In the following equation (39), GB (D), G 1 (D), and G 2 (D) are expressed by the following equations (40) to (42), respectively.
- GB [D) 1 + G + GB [l] D 2 + GB [2] D 3 (40)
- GI (D) G ⁇ [O] + G1 [I] D + G1 [2] D Z + G1 [3] D 3 (41)
- G 2 (D) G 2 [0] + G2 [l] D + G 2 [2] D 2 + G 2 [3] D 3 (42)
- a Massy type convolutional encoder for example, as shown in FIG. 2 1, and two shift register 207 2 07 2, three exclusive oR circuits 2 08 1, 208 2, 2 08 3, and 1 1 aND gates GB [0], GB [ 1], G1 [0], G1 [1], G1 [2], G2 [0], G2 [1], G2 [2], G3 [0], G3 [1] , G 3 [2], and performs a convolution operation with a coding rate of “3/3”.
- GI (D) GI [O] + GI [I] D + GI [2] D 2 (45)
- G2 ⁇ D G2 [0] + G2 [I] D + G2 [2] D '(46)
- This convolutional encoder has When inputting input data i. of bits, performs input data Isseki i. relative convolution of this, and outputs the operation result as a 4-bit output data Oo, 01, O 2, 0 3.
- a description of the trellis in this convolutional encoder is as shown in FIG.
- the label attached to each branch indicates the branch number.
- the relationship between the states before and after the transition and the input data / output data is as shown in Table 1 below.
- the number of states in the convolutional encoder shown in Fig. 22 is 16 and the trellis has a structure in which two paths reach from each state to the state at the next time. It has a structure having branches.
- the code information generation unit 151 outputs “1 bit” as input bit number information IN, “Posen-craft type” as type information WM, and “4 bits” as memory number information MN. Is used as branch input / output information BI 0 to generate an input / output pattern of each branch as shown in Table 1.
- the description of the trellis in this convolutional encoder is as shown in Fig. 25.
- the label attached to each branch indicates the branch number.
- the relationship between the state before and after the transition for this branch number and the input / output data / output data is as shown in Table 2 below.
- the state is a state in which the contents of the shift register 203 3 , the shift register 203 2 and the shift register 203 i are sequentially arranged, and “000”, “00 1 ,,, '(0 10 ,, “0 1 1 ,,," 100 ", '(101 ,,," 111 “,” 111 “state numbers,” 0 ",” 1 “,” 2 “,” 3 “,” 4 ",” 5 “,” represents a 6 ",” 7 ".
- the input data / output data, i l5 i./0 2, 0 i , in 0. is there. 9i / Jz.sz.dTHd Fiber O; AV:
- the number of states in the convolutional encoder shown in FIG. 24 is 8, and the trellis has a structure in which four paths reach from each state to the state at the next time. Will have.
- the code information generation unit 151 generates “2 bits” as input bit number information IN, “Posen-craft type” as type information WM, and “3” as memory number information MN. Is used as branch input / output information BI 0 to generate an input / output pattern of each branch as shown in Table 2.
- FIG. 26 when the three AND gates GB [2], G 1 [2], and G 2 [1] are connected as the Massey type convolutional encoder shown in FIG. 20, as shown in FIG. three shift Torejisu the evening 205 20 5 2 2 0 5 3, is considered to have a two exclusive OR circuits 2 0 6 2, 20 6 3.
- This convolutional encoder has a 2-bit input data i. , Enter it, these input data overnight i. , I!, Perform a recursive systematic convolution operation, and output the operation result as 3-bit output data 0. , 0 ,, 0 2 are output.
- the number of states in the convolutional encoder shown in FIG. 26 is 8, and the trellis has a structure in which four paths reach from each state to the state at the next time. It will have branches.
- the code information generation unit 151 outputs “2 bits” as input bit number information IN, “Massie type” as type information WM, and “3” as memory number information MN. Then, an input / output pattern of each branch as shown in Table 3 is generated as the branch input / output information BI0.
- the Massey-type convolutional encoder shown in FIG. 21 includes six AND gates GB [1], G1 [0], G1 [1], G1 [2], G2 [ 0], Tying G 3 [0], as shown in FIG. 28, two shift Torejisu evening 207 207 2, can be considered to have a three exclusive OR circuits 208 208 2, 208 3.
- the convolutional encoder inputs 3-bit input data i 0, ii, i 2 , these input data i. , Ii, i2, perform a recursive systematic convolution operation, and output the operation result to a 3-bit output data. , 01, 0 2 are output.
- the number of states in the convolutional encoder shown in Fig. 28 is 4, and the trellis has a structure in which four sets of parallel buses reach from each state to the state at the next time. Will have.
- the code information generation unit 151 outputs “3 bits” as input bit number information IN, “Massie type” as type information WM, and “2” as memory number information MN.
- An input / output pattern of each branch as shown in Table 4 is generated as input / output information BI0.
- the code information generation circuit 151 generates code information corresponding to the element encoder in the encoding device 1.
- the code information generation circuit 151 calculates the input / output patterns of all the branches on the trellis according to the code, and generates the branch input / output information B I0, which will be further described later.
- the code information generation circuit 15 1 converts the generated input bit number information IN into the termination information generation circuit 153, the received value and prior probability information 154, the Ia calculation circuit 156, the Ia distribution circuit 157, I
- the calculation circuit 158, the I-calculation circuit 159, the soft output calculation circuit 161 and the received value or prior probability information separation circuit 162 are supplied to the hard decision circuit 165.
- the code information generation circuit 155 supplies the generated type information WM to the Ia calculation circuit 156, the Ia distribution circuit 157, the Ih calculation circuit 158 and the I? Calculation circuit 159. . Further, the code information generation circuit 15 1 divides the generated memory number information MN into a termination information generation circuit 15 3, an I distribution circuit 1 57, an I calculation circuit 1 58, an I / 5 calculation circuit 1 59, And supplied to the soft output calculation circuit 16 1. Furthermore, the code information generation circuit 15 1 supplies the generated branch input / output information B I0 to the I distribution circuit 157 and the soft output calculation circuit 16 1. Further, the code information generation circuit 15 1 supplies the generated valid output position information PE to the internal erasure information generation circuit 152.
- the internal erasure information generation circuit 152 converts the puncture pattern and the effective output position based on the erasure information TERS supplied from the outside and the effective output position information PE supplied from the code information generation circuit 151.
- Internal erasure position information I ERS indicating a position where there is no code output obtained in consideration of the whole is generated.
- the internal erase information generation circuit 15 2 is realized as having four OR gates 2 1 1 2 1 1 2 1 1 1 1 3 and 2 1 1 4 as shown in FIG. 30, for example. Can be
- OR gates 2 1 1 2 1 1 2 , 2 1 1 3 , 2 1 1 4 are data obtained by inverting the erasure information TERS and the effective output position information PE supplied from the sign information generation circuit 15 1 And the logical sum of The OR gates 2 1 1 2 1 12, 2 1 1 3 , 2 1 1 4 supply the obtained OR to the received value and prior probability information selection circuit 154 as the internal erase position information IERS.
- the internal erasure information generation circuit 1 5 2 by taking the logical sum by OR gate one DOO 2 1 1 2 1 1 2, 2 1 1 3 2 1 1 4, the position where the code output is not present
- the internal erase position information IERS shown is generated.
- the termination information generation circuit 153 includes termination time information T TNP and termination state information T TNS supplied from outside, input bit number information IN and memory number information MN supplied from the code information generation circuit 151. Based on, the termination information in the encoding device 1 is generated. More specifically, the termination information generation circuit 153 determines the termination time in the encoder 1 based on the termination time information T TNP, termination state information T TNS, input bit number information IN, and memory number information MN. , And termination state information TSM indicating the termination state.
- the termination information generating circuit 153 includes a plurality of registers 2 121, 2 122, 2 1 2 3 , 2 1 2 4 , 2 1 2 5 , 2 1 2 6 , Multiple selectors 2 1 3 1, 2 1 3 2 , 2 1 3 3 , 2 1 3 "2 1 3 2 1 3 6 , 2 1 3 7 , 2 1 3 8 , 2 1 3 9 and AND gate 2 1 4 can be realized.
- Register 2 1 2 It is a termination time information T TNP supplied from outside and held 1 only the clock, and supplies the held termination time information TTNP to register 2 1 2 2 and selector 2 1 3 3.
- Register 2 1 2 2 is Register 2 1 2!
- Register 2 1 2 3 a termination time information T TNP supplied from the register 2 1 2 2 held by one clock and supply the held termination time information TTNP to the selector 2 1 3 5.
- Regis evening 2 1 24 holds termination state information T TNS supplied from outside by one clock, and supplies the held termination state information TTNS to register 2 1 2 5 and selectors evening 2 1 3 6.
- the register 2 1 2 5 holds the termination state information T TNS supplied from the register 2 1 2 4 for only one clock, and stores the held termination state information T TNS in the register 2 1.
- Register 2 1 2 6 register 2 1 2 5 termination state information ⁇ ⁇ S supplied from holding only one Kurodzuku, retained termination state information T TN S selector 2 1
- the selector 213 Based on the input bit number information IN, the selector 213 includes, for example, information indicating that the number of memories of the element encoder in the encoding device 1 is “1” among the memory number information MN. Select either one of the information indicating that the number of memories is "2". Specifically, for example, when the number of input bits in the encoding device 1 is “1”, the selector 213 selects information indicating that the number of memories is “1”. Selector 2 1 3! Supplies the selector evening 2 1 3 3 as a control signal for selecting the Isseki de selected.
- Information selector 2 1 3 2 based on the input bi Uz betting amount information IN, indicating that the Chi sac memory number information MN, for example, the number of memory elements encoder in the encoder 1 is "2" And information indicating that the number of memories is "3". Specifically, the selector 2 1 3 2, for example, when the number of input Kabi' DOO in the encoder 1 is "1", selects the information indicating that the number of memory is "2". Selector 2 1 3 2 selector evening supplies the 2 1 3 4 as a control signal for selecting the selected data.
- Selector 2 1 3 based on Isseki de selected by the selector 2 1 3 i, register 2 1 2, the termination time information T TNP value supplied from among the data is "1" , Select either one. Specifically, when the number of memories of the element encoder in the encoding device 1 is “1”, the selector 2 1 3 3 sets the register 2 1 2! Selects TTNP, the termination time information supplied from TTNP. Selector 2 1 3 3 was selected de - supplying data to the AND gate 2 1 4. Based on the data selected by the selector 2 13 2 , the selector 2 1 3 4 outputs the termination time information T TNP supplied from the register 2 1 2 2 and the data having the value “1”. Select one of them. Specifically, the selector 2 1 3 4
- the termination time information T TNP supplied from the register 2 1 2 2 is selected.
- Selector 2 1 3 4 supplies the Isseki de selected AND gate 2 14.
- Selector 2 1 3 5 on the basis of the number of memories information MN, out of the termination time information T TNP value supplied from the register evening 2 1 2 3 is "1" data, selects either. Specifically, the selector 2 1 3 5 selects the termination time information T TNP supplied from the register 2 1 2 3 when the number of memories of the element encoder in the encoding device 1 is “3”. I do. Selector 2 1 3 5 ANDs selected data with AND gate 2
- Selector 2 1 3 6 based on the number of memories information MN, registers 2 1 2 4 termination state information T TN S value supplied from Out of the de Isseki is "0", selects either I do. Specifically, the selector 2 13 ⁇ selects the termination state information TTNS supplied from the register 2 1 2 4 when the number of memories of the element encoder in the encoding device 1 is “1”. I do. Selector 2 1 3 6 supplies the selected data to the selector evening 2 1 3 8.
- Selector 2 1 3 7 on the basis of the memory number information Myunyu, register 2 1 2 5 termination state information T TN S value supplied from the "0" of the de one data is, selects either I do. Specifically, the selector 2 1 3 7 selects the termination state information TTNS supplied from the register 2 1 2 5 when the number of memories of the element encoder in the encoding device 1 is “2”. I do. Selector 2 1 3 7 supplies the selected data to the selector evening 2 1 3 8.
- Selector 2 1 3 8 on the basis of the memory number information Myunyu, among termination state information T TN S value supplied from the register evening 2 1 2 e is the data is "0", selects either . Specifically, when the number of memories of the element encoder in the encoding device 1 is “3”, the selector 2 1 3 8 outputs the termination state information T TNS supplied from the register 2 1 2 6. select. Selector 2 1 3 8 selects selected data Evening 2 1 3 8 supply.
- Selector 2 1 3 9 based on the input bit number information IN, data supplied from the final sintered state information T TNS supplied from the outside, the selector 2 1 3 6 2 1 3 7, 2 1 3 8 Select either one of Selector 2 1 3 9, supplied to the reception data and the delay storage circuit 1 5 5 the selected data as the termination state information T SM.
- AND gate 2 14 takes a termination time information T TNP supplied from outside, a logical product of the Isseki de supplied from selector evening 2 1 3 3 2 1 34, 2 1 3 5.
- the AND gate 2 14 supplies the obtained logical product to the reception data and delay storage circuit 1 55 as termination time information TPM.
- Such termination information generation circuit 153 grasps the termination period based on the number-of-memory information MN, and selects a data selection according to the termination period by selectors 2 1 3 3 , 2 1 3 4) 2 by performing the 1 3 5 2 1 3 6 2 1 3 7, 2 1 3 8, it is possible to generate termination information for any termination period.
- the termination information generation circuit 153 outputs the number of input bits as termination information. The termination state is specified by generating minute information for the termination period.
- the termination information generation circuit 153 is used when the element encoder in the encoding device 1 is other than a Posencraft type convolutional encoder such as a matsy type. Specifies the end state in one time slot by generating information indicating the end state in one time slot as the end information.
- the received value and prior probability information selection circuit 154 is provided for decoding an arbitrary code, as described later.
- the received value and prior probability information selection circuit 154 includes the received value format information CRT Y supplied from the control circuit 60, the input bit number information IN supplied from the code information generation circuit 151, and the external supply. Based on the prior probabilistic information erasure information TEAP and the internal erasure position information IERS supplied from the internal erasure information generation circuit 152, the input decoded reception value TSR and external information or interleave data TEXT are used. Among them, the information necessary for performing soft output decoding is selected. Also, the received value and prior probability information selection circuit 154 has an internal eraser as described later.
- a position where no code output exists is replaced with a symbol having a likelihood of "0". That is, the reception value and prior probability information selection circuit 154 determines that the probability that the bit corresponding to the position where no code output exists is “0” or “1” is “1/2”. It outputs the information that you want.
- the reception value and prior probability information selection circuit 154 includes, for example, a method in which the decoded reception value TSR is composed of four systems of decoded reception values TSR0, TSR1, TSR2, and TSR3, and Or, if the TEXT is composed of three sets of external information or interleaved data TEXT0, TEXTl, and TEXT2, for example, as shown in Fig.
- the selector 215 selects one of the decoded received value TSR0 and the external information or in-out / leave-out / TEXT0 based on the received value format information CRTY. Specifically, the selector 2 1 5! If the received value format information CRTY indicates external information, the external information or the interleaved data TEXT 0 is selected. Selector 2 1 5, supplies the Isseki de selected in the selector 2 1 5 8. Selector 2 1 5 2, based on the received value type information CRT Y, and the decoded received value T SR 1, out of the external information or inter one Ribude Isseki TEXT 1, either to select.
- the selector 2 1 5 when the received value type information CRT Y were indicates the external information, the external information or inter one leave data TEXT 1 to select.
- Selector 2 1 5 2 supplies Isseki de selected in the selector 2 1 5 9.
- Selector 2 1 5 3 based on the received value type information CRT Y, and the decoded received value T SR 2, among the extrinsic information or interleaved de Isseki TEXT 2, either to select.
- the selector 2 1 5 3 when the received value type information CRT Y were indicates the external information, the external information or fin evening one leave data TE XT 2 to select.
- the selector 2 1 5 3 selects the selected data with the selector 2 1 5! . To supply.
- Selector 2 1 5 based on the received value type information CRT Y, external information or in the evening —Select either leave data T EXT 0 or prior probability information whose value is “ ⁇ ”. Specifically, the selector 2 1 5 4, when the received value type information CRT Y were indicates the external information, the value you select a priori probability information is "0". Selector 2 1 5 4 supplies the selected data to the selector 2 1 51 2.
- the selector 2 1 5 5 selects either the external information or the in-leave data TEXT 1 and the prior probability information whose value is “0”, based on the received value format information CRTY. Specifically, the selector 2 1 5 5, when the received value type information CRT Y were indicates the external information, the value you select a priori probability information is "0". Selector 2 1 5 5 supplies Isseki de selected in the selector 2 1 5 3.
- Selector 2 1 5 beta based on the received value type information CRT Upsilon, the external information or inter one Ribude Isseki TEXT 2, of a priori probability information whose value is "0", selects either. Specifically, the selector 2 1 5 6, when the received value type information CRTY was shows the external information, the value you select a priori probability information is "0". The selector 2 15 ⁇ selects the selected data with the selector 2 15! Supply 4
- Selector 2 1 5 7 based on the received value type information CR T Upsilon, in the internal erasure locator information I ER S, for example, one of the output bits that are output from the element encoder in the encoder 1 Either the information indicating that the symbol does not exist or the information indicating that the second symbol does not exist is selected.
- selector 2 1 5 7 when the coding apparatus 1 were those indicated in the received value type information CR TY that it is not intended to perform coding by TT CM or SCT CM is 2 Shinporu Select information that indicates that no eyes are present.
- Selector 2 1 5 7 supplies to the selector 2 1 5 9 as a control signal for selecting the selected data.
- selecting operation of the selector 2 1 5 7 is due to the erase operation in a coding apparatus 1 was used to encode by TT CM and SCT CM. That is, the erase operation when the code KaSo location 1 was used to encode by TT CM and SCT CM, since becomes to erase the symbol of the in-phase and quadrature components both cell Lek evening 2 1 5 7 Means to select information indicating that the second symbol does not exist.
- Selector 2 1 5 based on the internal erasure position information I ER S, selector 2 1 5 i Either the data supplied from the server or the information whose value is "0" is selected. Specifically, the selector 2 1 5 8, there is shown that the internal erasure locator information I ERS does not exist 1 Shinporu th among the output bits outputted from the element encoder in the encoder 1 In this case, select the information whose value is "0".
- the data selected by the selector 2 1 5 9, selector 2 1 58, 2 1 510, 2 1 514, 2 1 5 IS, 2 1 5 16 bundles with the data supplied from Nerare, selected received value and a priori probability
- the received data and the delay storage circuit 155 are supplied as information RAP.
- Selector 2 15 i Based on the internal erasure position information I ERS, the data supplied from the selector 2 1 5 3, of the value is "0" information, selects either. Specifically, the selector 2 1 5! . If the internal erasure position information IERS indicates that the third symbol of the output bits output from the element encoder in the encoding device 1 does not exist, the value is " Select information that is 0 ". Selector 2 1 5,. Data is selected by the selector 2 1 5 8, 2 1 5 9, 2 1 514, 2 1 5 is, 2 1 5 1 ⁇ bundled together Isseki de supplied from the selection received value and a priori probability information It is supplied to the received data and delay storage circuit 155 as RAP.
- the selector 215H selects one of the decoded received value TSR3 and the information whose value is "0" based on the internal erasure position information IERS. Specifically, the selector 215H indicates that the internal erasure position information IERS does not exist in the fourth symbol of the output bits output from the element encoder in the encoder 1. If it is, select the information whose value is "0". The selector 215 n supplies the selected data to the selectors 215, 5 .
- Selector 2 1 5 based on the a priori probability information erasure information TE AP, selects and de-Isseki supplied from the selector 2 1 5 4, of the value is "0" information, either . Specifically, if the prior probability information erasure information TEAP indicates that the punctured information TEAP has been punctured, the selector 2 15 12 selects information having a value of “0”. The selector 2 15 12 supplies the selected data to the selectors 2 15 15 and 2 15 16 .
- Selector 2 1 5 13 based on the a priori probability information erasure information TEAP, the data supplied from the selector 2 1 5 5, of the value is "0" information, selects either. Specifically, when the prior probability information erasure information TEAP indicates that the punctured information TEAP has been punctured, the selector 2 15 13 selects information having a value of “0”. Selector 2 1 5 13, you supply Isseki de selected in the selector 2 1 5 16.
- Selector 2 1 5 14 based on the a priori probability information erasure information TEAP, the data supplied from the selector 2 1 5 6, of the value is "0" information, selects either. Specifically, the selector 2 1 5 14, when a priori probability information erasure information TE AP was indicates that it is Pankuchiya, the value selects information is "0". Data selected by the selector 2 1 5 14 bundled together with the data supplied from the selector 2 1 5 8, 2 1 5 9, 2 1 5 10, 2 1 5 IS, 2 1 5 16, selected received value and It is supplied to the received data and delay storage circuit 155 as prior probability information RAP.
- the selector 215 15 receives one of the data supplied from the selector 215 and the data supplied from the selector 215> 2 based on the input bit number information IN. Select Specifically, the selector 2 1 5 15, the element encoder code rate in the encoder 1 is represented by "1 / n", and the input that the input number of bits is "1" If the bit number information IN indicates this, the data supplied from the selector 215 u is selected. Data selected by the selector 2 1 51 5 is supplied from the selector 2 1 5 8, 2 1 5 9, 2 1 5 10, 2 1 5 Eta, 2 1 5 16 The data is bundled together with the data and supplied to the received data and the delay storage circuit 155 as the selected reception value and the prior probability information RAP.
- the selector 215 1 ⁇ selects one of the data supplied from the selector 215 12 and the data supplied from the selector 215 13 based on the input bit number information I ⁇ . Select one. Specifically, the selector 215 1 ⁇ determines that the coding rate of the elementary encoder in the coding device 1 is represented by “1 / ⁇ ” and that the number of input bits is “1”. If the input bit number information I indicates, the data supplied from the selector 215 i 2 is selected.
- the received value and the prior probability information RAP are supplied to the received data and delay storage circuit 155 as RAP.
- Such a received value and prior probability information selection circuit 154 uses the selector 2 15 2 1 52, 2 1 5 3 , 2 1 5 4 , 2 1 5 5 , 2 1 5 6 to obtain the decoded received value TSR and external information.
- the decoded received value TSR and the extrinsic information sentence can be switched between the interleaved data TEXT and the interleaved data TEXT as the code likelihood, and the soft output decoding is performed.
- Information to be input for the purpose can be appropriately selected.
- the reception data and delay storage circuit 155 has, for example, a plurality of banks of RAMs, a control circuit, and a selection circuit.
- the reception data and delay storage circuit 155 includes the termination time information TPM and termination state information TSM supplied from the termination information generation circuit 153, and the selected reception value supplied from the reception value and prior probability information selection circuit 154. And prior probability information RAP.
- the reception data and delay storage circuit 155 selects predetermined information from the stored termination time information TPM and termination state information TSM by a selection circuit.
- the termination information TAL, I used in the calculation circuit 158 and the termination information TB0, TBI used in the calculation circuit 159 are output. Ending The information TAL is supplied to the I-calculation circuit 158 as termination information TALD after a predetermined delay.
- the termination information TB 0 and TB 1 are supplied to the 1/5 calculation circuit 159 as termination information TB 0 D and TB 1 D after a predetermined delay, respectively.
- the received data and delay storage circuit 155 selects predetermined information from the stored selected reception value and prior probability information RAP by the selection circuit, and calculates Received data DA used in circuit 158,? Output as two systems of received data DB 0, DB 1 used in calculation circuit 159.
- the received data DA is supplied to an I-a calculation circuit 156 and, after being subjected to a predetermined delay, is supplied to a received value or prior probability information separation circuit 162 as delayed reception data DAD. Further, the received data DB 0 and DB 1 are supplied to the I-a calculation circuit 156 respectively.
- the element decoder 50 performs a so-called sliding window process known as a method for processing continuous data.
- the received data and delay storage circuit 155 and an I? As a method of memory management in the storage circuit 160, the one described in International Publication No. WO99 / 62183, which the applicant of the present invention has already applied for an international patent, is employed. That is, the element decoder 50 reads out the reception data divided by a predetermined truncation length from the reception data and the delay storage circuit 155 in a simplified manner, and the log likelihood By storing I?, Memory management is performed such that the logarithmic soft output I ⁇ is finally obtained in the original chronological order.
- the element decoder 50 has the international publication number WO 9 9/6 2 183 As described in the official gazette, instead of calculating the log likelihood Ia and then performing the memory management, the received data is stored in the received data and the delay storage circuit 155, and then an appropriate The received data is read out under memory management, and the log likelihood Ia is calculated.
- the reception data and delay storage circuit 155 can also store delay data as described later. That is, the storage circuit for received data and delay 155 receives the received value TR and the edge signal TEILS supplied from the edge detection circuit 80. And delay the same time as the processing time required by the soft output decoding circuit 90. Received data and delay storage circuit 1 5 5 supplies the delayed received value PDR obtained by delaying the received value TR, to the selector 1 2 0 3, 1 2 0 6 as delayed received value SDR. Further, the received data and delay storage circuit 1 5 5 supplies the delayed edge signal PD IL obtained by delaying the edge signal TEILS, the selector 1 2 0 5 as a delay edge signal SD ILS.
- the I-key output circuit 156 calculates the log likelihood I ⁇ ⁇ using the received data DA, DB 0, and DB 1 supplied from the received data and delay storage circuit 155. Specifically, based on the notation described at the beginning of “2.”, the Ia calculation circuit 156 performs the calculation shown in the following equation (48) for each received value y t , and at each time t Calculate the log likelihood Ia. Note that s gn shown in the following equation (48) is a constant indicating a sign for distinguishing between positive and negative, that is, either “+1” or “11”.
- the element decoder 50 is configured as a system that handles only negative values as log likelihood, this constant sgn takes “+1”, and the element decoder 50 handles only positive values as log likelihood. If configured as a system, take "1-1". That, I y calculating circuit 1 5 6, for each received value y t, positive or negative identification log likelihood I ⁇ or probability ⁇ were log notation the probability ⁇ determined by the sign of the output pattern and the received value by logarithmic notation Calculate the log-likelihood I with inverted sign.
- the Ia calculation circuit 156 generates the received value format information supplied from the control circuit 60.
- the log likelihood Iy is calculated based on the supplied input bit number information IN and the type information WM.
- the Ia calculation circuit 156 supplies the calculated log likelihood Ia to the Ia distribution circuit 157. That is, the Ia calculation circuit 156 supplies the log likelihood Ia used in the I calculation circuit 158 to the Ia distribution circuit 157 as a log likelihood GA, and uses it in the I? Calculation circuit 159.
- the log likelihood Ia is supplied to the Ia distribution circuit 157 as log likelihoods GB0 and GB1.
- the Ia calculation circuit 156 uses the log likelihood I.sub.0 used to calculate the log likelihood I.sub.0 of the two systems of log likelihood 1.sub.0. calculating a ⁇ 1? 0 for I ⁇ calculating circuit 2 20, and, as I ⁇ calculating circuit 2 2 0 2 for I / 51 for calculating a log likelihood I ⁇ used to calculate the log likelihood I beta 1 it can be implemented as being closed and logarithmic likelihood I I ⁇ calculating circuit I shed for calculating a ⁇ 2 2 0 3 used for calculating the log likelihood I monument.
- I ⁇ calculating circuit 2 2 0 "for I ⁇ calculating circuit 2 20 2 and I I ⁇ calculating circuit 2 20 3 for non for these I 50 is realized by the same configuration with the data to be input is different only because it can, in this case, performs the description for I 0 I ⁇ calculation circuit 22 0 only, I; nowadays description 51 for I ⁇ calculating circuit 2 20 2 and I shed for I ⁇ calculating circuit 2 2 0 3 is shown Omitted with
- Ia calculation circuit for I? 0 220! Has an information / sign Ia calculation circuit 221 and an Ia normalization circuit 222.
- the information and code I key calculation circuit 221 receives the reception data DB0 including the reception value and the prior probability information, and receives the reception value format information CRTY, the prior probability information format information CAP P, and the signal point arrangement. Based on information CSIG and input bit number information IN, log likelihood Ir for all possible input / output patterns or log likelihood Ir for at least a part of input / output patterns is calculated.
- the information and code I key calculation circuit 2 21 transmits the prior probability information and the so-called communication
- the sum with the road value is calculated as the log likelihood Ia.
- the information / code I-key calculation circuit 2 21 If coding by M is performed, the log likelihood I a is calculated by calculating the inner product of the input received data DB 0. This is because the Euclidean distance on the I / Q plane is the log likelihood Ia, but in the case of the PSK modulation method, the transmission amplitude of the output from the encoder takes a constant value, so Finding the distance is equivalent to finding the inner product.
- the Ia normalization circuit 222 performs normalization to correct the bias of the distribution of the calculation result by the information / code Ia calculation circuit 222.
- the Ia normalization circuit 2 2 2 corresponds to the information ′ code Ia calculation circuit 2 2 1 having a maximum probability value among a plurality of log likelihood Iers calculated by the information I code calculation circuit 2 2 1.
- a predetermined operation is performed on each log likelihood so that the log likelihood matches the log likelihood corresponding to the maximum value of the probability that can be taken. That is, when the element decoder 50 treats the log likelihood as a negative value, the Ia normalization circuit 2 2 2 2 outputs the plurality of log likelihoods I calculated by the information / code Ia calculation circuit 2 2.
- the Ia normalization circuit 2 2 2 2 outputs a plurality of log likelihoods I calculated by the information / code Ia calculation circuit 2 2 1.
- the Ia normalization circuit 2 2 2 clips the normalized log likelihood Ia according to the required dynamic range, and supplies it as the log likelihood GB 0 to the Ia distribution circuit 1 57. .
- Such an I 0 Ia calculation circuit 220! Calculates a log likelihood Ia used to calculate the log likelihood I ⁇ 0, and supplies the log likelihood GB0 to the Ia distribution circuit 157.
- I ⁇ calculation circuit 2 2 0 2 for I? 1 is, I? 0 for I ⁇ calculation circuit 2 2 0!
- the received data D ⁇ ⁇ ⁇ 1 is input instead of the received data D ⁇ 0 input to the, and the same processing as the I / O calculation circuit 220 for 1? 0 is performed.
- the I a calculation circuit for I? 1 2 2 2 The log likelihood Ia used to calculate the degree I-1 is calculated and supplied to the Ia distribution circuit 157 as the log likelihood GB1.
- I ⁇ calculating circuit 2 2 0 3 for non-I instead of receiving de Isseki DB 0 which is input to the I ⁇ calculating circuit 2 20, for I 0, enter the received de Isseki DA, 1 ? I output circuit for 0 22 0! The same processing is performed.
- I ⁇ calculation circuit 22 0 3 for non-I calculates log likelihood I ⁇ used to calculate the log likelihood I alpha, and supplies the I ⁇ distribution circuit 1 57 as a log-likelihood G Alpha.
- Such an Ia calculation circuit 156 uses the received data DA, DB0, and DB1 to generate log likelihood GA, GB0, and GB1 calculated as the log likelihood Ia.
- Log likelihood GA, GB O, and GB 1 are supplied to the I key distribution circuit 157.
- the Ia distribution circuit 157 distributes the log likelihood GA, GBO, and GB1 supplied from the Ia calculation circuit 156 according to the code configuration. That is, the Ia distribution circuit 157 distributes the log likelihood GA, GB0, and GB1 so as to correspond to the branches on the trellis according to the code configuration.
- the Ia distribution circuit 157 includes the generation matrix information CG supplied from the control circuit 60, the input bit number information IN, the type information WM, and the memory number information MN supplied from the code information generation circuit 151.
- the log likelihood GA, GB O, and GB 1 are distributed based on the branch input and output information BI 0.
- the Ia distribution circuit 157 when decoding a code having a parallel path on the trellis, the Ia distribution circuit 157 also has a function of bundling these parallel paths.
- the Ia distribution circuit 157 supplies the log likelihood Ia obtained by the distribution to the I calculation circuit 158 and the I ⁇ calculation circuit 159. That is, the Ia distribution circuit 157 supplies the log likelihood Ia used in the I subtraction circuit 158 to the I calculation circuit 158 as a log likelihood DGA, and the I? The log likelihood Ia to be used is supplied to the 1? Calculation circuit 159 as log likelihood DGB O, DGB 1. In addition, as will be described later, the Ia distribution circuit 157 supplies the logarithmic likelihood Ia obtained in a state in which the parallel paths are not bundled to the Ih calculation circuit 158 as a logarithmic likelihood D GAB.
- the I / A distribution circuit 157 includes a branch input / output information calculation circuit 2 23 that calculates input / output information of a branch on a trellis according to a code configuration, Of the two sets of log likelihood I? 0, I 50 distribution circuit 2 2 4 for distributing log likelihood I a used to calculate log likelihood I 50, and log likelihood The I 1 distribution circuit 2 2 4 2 for distributing the log likelihood I ⁇ ⁇ ⁇ used to calculate the degree I?
- the branch input / output information calculation circuit 222 forms a code based on generator matrix information CG, input bit number information IN, type information WM, memory number information MN, and branch input / output information BI0. , And calculates branch input / output information along the reverse order of the time axis of the branch on the trellis corresponding to the code configuration.
- the branch input / output information calculation circuit 2 23 uses the calculated branch input / output information BI as an I 0 distribution circuit 2 2 4! And supplies the I ⁇ distribution circuit 2 2 4 2 for I 5 1.
- I? Distribution circuit for I? 0 2 2 4! Receives the log likelihood G B 0 and performs distribution according to the code configuration based on the branch input / output information B I.
- I? Distribution circuit for I? 0 2 2 4! Is a logarithmic likelihood PGB0 obtained by distributing the parallel path processing circuit for I? To be supplied.
- the I 1 distribution circuit 2 2 4 2 When the log likelihood GB 1 is input, the I 1 distribution circuit 2 2 4 2 performs distribution according to the code configuration based on the branch input / output information BI.
- the I /? 1 I a distribution circuit 2 2 4 2 supplies the log likelihood PGB 1 obtained by the distribution to the I? 1 parallel path processing circuit 2 25 2 .
- the I utilization I distribution circuit 2 2 4 3 When the logarithmic likelihood GA is input, the I utilization I distribution circuit 2 2 4 3 performs distribution according to the code configuration based on the branch input / output information BI0. I I ⁇ distributing circuit 2 2 4 3 for non supplies the log likelihood PGA obtained by arranging minute parallel path processing circuit 2 2 5 3 for non I. In addition, the I utilization I distribution circuit 2 2 4 3 uses the log likelihood PGA obtained by the distribution as a logarithm It is supplied as likelihood DGAB to the I calculation circuit 158.
- the parallel path processing circuit 22 5 i for I 0 the parallel path processing circuit 22 5 i for I 0
- the likelihood PGB 0 is bundled and output as the log likelihood DGB 0, that is, the log likelihood Ia used to calculate the log likelihood 1 to 0.
- the parallel path processing circuit 225 for I-0 selects the log likelihood DGB0 to be output based on the input bit number information IN.
- the parallel path processing circuit for I? 0 2 2 5! As shown in Fig. 35, the log-to- sum operation circuit 2 26 n for the parallel path of the maximum number of states of the code to be decoded and a selector for selecting 2 to 1 2 27
- the parallel path processing circuit for I? 0 2 2 5! Is a code represented by a trellis having a maximum of 32 branches and having a maximum of 4 states among codes having a parallel path on the trellis, and more specifically, a code having a maximum of 4 states.
- a selection control signal generation circuit 232 that generates a control signal for controlling the selection operation by the selectors 230, 231, and 233, and a correction term value in so-called log-sum correction are stored as a table. It has a look-up table 234 composed of ROM (Read Only Memory) and the like, and an adder 235. Among these units, the difference units 229 1, 229 2 , the selectors 230 231, and the selection control signal generation circuit 232 constitute a comparison and absolute value calculation circuit 228.
- the comparison and absolute value calculation circuit 228 compares the magnitudes of the two input data, and calculates the absolute value of the difference value between the two data.
- the differentiator 2 2 9 i is composed of two sets of log likelihoods I log, PGB 0, two log likelihoods I log PG 0 0 and log likelihood PG Take the difference from 0 1. Strictly speaking, the differencer 2 2 9! Assuming that each of the log likelihoods PG00 and PG01 is composed of, for example, 9 bits, the most significant bit of the lower 6 bits of the log likelihood PG00 is "1". The difference between the one with "" and the one with "0” added to the most significant bit of the lower 6 bits of the log likelihood PGO1 is calculated. Differentiator 2 2 9! Supplies the calculated difference value DA 1 to the selector 230 and the selection control signal generation circuit 232.
- Differentiator 2 2 9 2 takes a difference between the log likelihood PGO 1 and logarithmic likelihood PGO 0. Strictly speaking, differentiator 2 2 9 2 log likelihood PG 0 0, PG 0 1 is, it it, for example 9 When made of bi Uz bets, the lower six bits of the log likelihood PG 0 1 The difference between the one with "1" added to the most significant bit of the data overnight and the one with "0" added to the most significant bit of the lower 6 bits of the log likelihood PG 00 Take. Differentiator 2 2 9 2 supplies a difference integral value DA 0 calculated for the selector 2 3 0 and the selection control signal generating circuit 2 3 2.
- the selector 2 30 Based on the control signal SL 1 supplied from the selection control signal generation circuit 23 2, the selector 2 30 generates a difference value DA 1 supplied from the difference device 2 29 i and a difference value DA 1 supplied from the difference device 2 29 2 Among the supplied difference values DA 0, the one with the larger value is selected.
- the selector 230 supplies the selected data CA to the selector 231.
- the selector 231 between the data CA supplied from the selector 230 and the data having a predetermined value M, based on the control signal SL2 supplied from the selection control signal generation circuit 2332. , Select either one. More specifically, since the value of the correction term for the difference value supplied as the data CA has the property of asymptotically approaching a predetermined value, the selector 231 sets the value of the data CA to a predetermined value. If the value M is exceeded, select the data having the predetermined value M. The selector 231 supplies the data DM obtained by the selection to the look-up table 234.
- the selection control signal generation circuit 232 controls the selection operation by the selectors 230, 233 based on the log likelihoods PG00, PG01 and the difference values DA1, DA0. And a control signal SL2 for controlling the selection operation by the selector 231.
- the selection control signal generation circuit 2 3 2 generates control signals SL 1 and SL 2 that divide the upper bit and the lower bit of the metric based on the log likelihood PG 00 and PG 01 and indicate a decision sentence for selection However, this will be described later.
- Such a comparison and absolute value calculation circuit 228 calculates the absolute value of the difference value between the log likelihoods PG 00 and PG 01.
- the difference unit 229 Assuming that the log likelihood PG 00 and PG 01 each consist of, for example, 9 bits, the most significant bit of the data of the lower 6 bits of the log likelihood P GO 0 is And "0" added to the most significant bit of the lower 6 bits of the log likelihood PG01.
- the comparison and absolute value calculation circuit 22 8 is de one evening is supplied to the differentiator 2 29 2, the top-level bi Uz Doo data of the lower 6 bits of the log likelihood PG 00 "0" , And "1" is added to the most significant bit of the lower 6 bits of the log likelihood PG01. That is, the differentiator 2 291 229 2, log likelihood PG 0 0, PG 0 1 to the most significant bit of the lower bi Uz bets of "1" or "0" is assigned data is supplied.
- the magnitude of the log likelihoods PGOO and PG01 can be compared at a high speed, and the upper and lower bits of the metric are selected by the selection control signal generation circuit 232. This is related to creating a decision sentence for selection. This will be described later.
- the selector 233 selects a log likelihood PB 00 or PG 01 having a smaller value based on the control signal SL 1 supplied from the selection control signal generation circuit 232.
- the selector 233 supplies the selected data SPG to the adder 235.
- the look-up table 234 stores the value of the correction term in the 10 g-sum correction as a table.
- the look-up table 234 reads the value of the correction term corresponding to the value of the data DM supplied from the selector 231, from the table, and supplies it as data RDM to the adder 235.
- the adder 235 adds the data SPG supplied from the selector 233 and the data RDM supplied from the look-up table 234 to calculate a log likelihood Ir.
- the adder 235 selects the calculated log likelihood Ia as log likelihood PP GO 0. 2 to 27.
- the parallel path 10 g—s um arithmetic circuit 2 2 6 2 has the same configuration as the parallel path 1 og—s 11 m arithmetic circuit 2 2 6, and has two log likelihoods PG 0 corresponding to the parallel path. 2. Bundle PG03 and supply it to selector 227 as log likelihood PP GO1.
- the parallel path 10 g-sum arithmetic circuit 2 2 6 3 is a parallel path lo-sum arithmetic circuit 2 2 6! PG04 and PG05 corresponding to the parallel path are bundled and supplied to the selector 227 as the log likelihood PPG02.
- the 1 og-sum arithmetic circuit 2 2 16 for the parallel bus is 1 og-sum arithmetic circuit 2 2 6 for the parallel path!
- the configuration is similar to that of, and two log likelihoods PG030 and PG031 corresponding to the parallel path are bundled and supplied to the selector 227 as the log likelihood PPG15.
- the plurality of parallel bus 10 g-sum arithmetic circuits 2 26 ⁇ bundle the two log likelihoods corresponding to the parallel paths.
- the log likelihood PPG 0 0, PPG 01, PPG 0 2, ⁇ , PP G 15 obtained by bundling by the log- sum operation circuit 2 2 6 n for each parallel path is the log likelihood Supplied to the selector 227 as PPG.
- the selector 227 is based on the input bit number information I, and the I? 0 I distribution circuit 224! Of the log-likelihood P GB 0 supplied from the log- sum operation circuit for each parallel path and the log-likelihood P P Select one of them. Specifically, the selector 227 selects the log likelihood PPG when the element encoder in the encoding device 1 performs encoding in which a parallel path exists on the trellis. That is, here, the input bit number information IN is used as a control signal for controlling the selection operation by the selector 227. At this time, a control signal indicating whether or not the code has a parallel path on the trellis is input to the selector 227.
- the parallel path processing circuit for I 0 0 2 25 t When such a log path likelihood PGB 0 is input, the parallel path processing circuit for I 0 0 2 25 t generates a selector 2 27 if this log likelihood PGB 0 corresponds to the parallel path.
- the log likelihood PPG is selected, and the log likelihood PPG and the log metric PGB 0 that correspond to the higher metric are combined, and the log likelihood D GB 0 Is supplied to the I? Calculation circuit 159.
- the parallel path processing circuit 2 25 t for 1 to 0 uses the log likelihood PGB 0 as the log likelihood DGB 0 as it is. Output.
- the parallel bus processing circuit 2 25 2 for I-1 has the same configuration as the parallel path processing circuit 2 25 t for I / 50, detailed description is omitted, but log likelihood PGB 1 is input. Then, if the log likelihood P GB 1 corresponds to the parallel path, the log likelihood P GB 1 is bundled, and the log likelihood D GB 1, that is, the log likelihood I ⁇ 1 is calculated. Is supplied to the 1? Calculation circuit 159 as the log likelihood Ia used for the calculation. Also, if the input log likelihood PGB 1 does not correspond to the parallel path, the parallel path processing circuit 2 2 5 2 for I ⁇ 1 uses this log likelihood P GB 1 as the log likelihood DGB 1 As it is, it is supplied to the 1/5 calculation circuit 159.
- the parallel path processing circuit 2 25 3 for I has the same configuration as the parallel path processing circuit 2 25 5 for I 0, detailed description is omitted, but the log likelihood PGA is input. Then, if this log likelihood PGA corresponds to the parallel path, the log likelihood PGA is bundled and the log likelihood DGA, that is, the log likelihood used to calculate the log likelihood I It is supplied to the I calculation circuit 158 as Ia. If the input log likelihood PGA does not correspond to the parallel path, the I parallel path processing circuit 225 3 sets the log likelihood PGA to the log likelihood DGA as it is. Calculating circuit 1 58
- Such an I distribution circuit 157 distributes the log likelihood GA, GB0, and GB1 according to the code configuration, and decodes a code having a parallel path on the trellis. At this time, these parallel paths are bundled, and the obtained log likelihood DGA, DGAB is supplied to the I-calculation circuit 158, and the obtained log likelihoods DGB0 and DGB1 are supplied to the 1-calculation circuit 159.
- the I-likelihood calculation circuit 158 calculates the log likelihood I using the log likelihoods DGA and DGAB supplied from the I-a distribution circuit 157. More specifically, based on the notation described at the beginning of "2.", the I-height calculation circuit 158 performs the operation shown in the following equation (49) using the log likelihood Ir, and calculates each time t Calculate the log likelihood I at. Note that the operator “#” in the following equation (49) indicates a so-called 10 g-sum operation, and the log likelihood when the input “0” transitions from state m to state m is This figure shows the 1 og-sum operation with the log likelihood at the time of transition from state m,, to state m at input "1".
- the I calculation circuit 158 performs the operation shown in the following equation (50), whereby the constant sgn becomes “1 1
- the logarithmic likelihood I at each time t is calculated by performing the operation shown in the following equation (51).
- the I-history calculation circuit 158 calculates, for each received value y t , the log-likelihood I that expresses the probability of reaching each state in chronological order from the coding start state in logarithmic order.
- the log likelihood I is calculated by inverting the sign of sign or probability by inverting the sign.
- the I-parameter calculation circuit 158 includes the generation matrix information CG supplied from the control circuit 60, the input bit number information IN supplied from the code information generation circuit 151, and the type information WM. Then, the log likelihood I is calculated based on the memory number information MN and the received data and the termination information TALD supplied from the delay storage circuit 155.
- the I-line calculation circuit 158 supplies the sum of the calculated log-likelihood I and the log-likelihood Ia to the soft output calculation circuit 161. That is, as will be described later, the I-height calculation circuit 158 does not output the calculated log-likelihood I as it is, but uses the log-likelihood Ia and log-likelihood I And outputs the sum as the data AG.
- the I-parameter calculation circuit 158 includes a control signal generation circuit 240 that generates a control signal and a state from each state on the trellis to a state at the next time.
- Addition / comparison / selection circuit 2 4 1 that performs addition and comparison (add co immediate are select) processing and processing to add a correction term by 1 og-sum correction for a code such that two paths arrive.
- Addition / comparison / selection processing and 10 g sum for a code such that four paths or eight paths depending on the code arrive from each state on the trellis to the state at the next time Add processing to add a correction term using Sodeshi It is realized as having an arithmetic and comparison circuit 242, an I + Ia calculation circuit 243 for calculating the sum of the log likelihood I and the log likelihood Ia, and a selector 244 for performing a 3: 1 selection. be able to.
- the control signal generation circuit 240 uses the generation matrix information CG, input bit number information IN, type information WM, and memory number information MN to generate four paths from each state on the trellis to the state at the next time. Then, the state of the transition source of the code that reaches the value is calculated and supplied to the addition / comparison / selection circuit 242 as the control signal PST.
- the addition / comparison / selection circuit 241 corrects the code such that two paths arrive from each state on the trellis to the state at the next time by addition / comparison / selection processing and 10-sum correction
- the log-sum operation is performed by adding a term.
- the addition / comparison / selection circuit 241 is a decoding target for a code in which two paths arrive from each state on the trellis to the state at the next time.
- 10 g s um operation circuit 245 n of the maximum value of the number of states of the code.
- the addition / comparison / selection circuit 24 1 decodes a code having a maximum of 16 states, and 16 10 g-sum operation circuits 2 451, 2452, 2453, ⁇ , 245 ! It shall have 6 .
- log-sum operation circuits 2451, 245 2 , 245 3 ,..., 245 16 have the log likelihood I of the branch corresponding to the output pattern on the trellis based on the transition on the trellis. And the log-likelihood I in the previous state in each state. That, 10 g- s um arithmetic circuit 245 24 5 2, 245 3, ...., 245 in the l [beta], respectively, of the log likelihood D GA, the edge corresponding to the output path evening one down the trellis log The one corresponding to the likelihood I a and the one corresponding to the log likelihood I in each state among the calculated log likelihoods AL one time ago are supplied.
- log- s um arithmetic circuit 245 24 5 2, 245 3, ⁇ ⁇ ⁇ , 245 i 6 it it determines the log likelihood I Monument in each state of the next time as log likelihood AL.
- log likelihood AL The distribution of the log likelihood AL to ⁇ differs depending on the code configuration.
- a selector (not shown) is used based on the memory number information ⁇ . Etc. are determined. This log likelihood A The distribution of L will be further described later.
- 10 g—sum arithmetic circuit 2 4 5! Are the three adders 2 4 6!, 2 4 6 2 j 2 4 9, the correction term calculation circuit 2 4 7 for calculating the value of the correction term in the 10 g-sum correction, and the selector 2 4 8 And I normalization circuit 250.
- the adder 2 46 i inputs the log likelihood DGA 00 of the log likelihood DGA and, among the log likelihoods AL calculated one time earlier, the one corresponding to the sign. Input as log likelihood A 0 and add these log likelihoods DGA 00 and A 0.
- the adder 2 4 6 "supplies the data AM0 indicating the sum of the log likelihood I obtained by the addition and the log likelihood I ⁇ to the correction term calculation circuit 2 4 7 and the selector 2 4 8 .
- the adder 2 4 6 2 inputs the log likelihood DGA 0 1 of the log likelihood DGA, and outputs the log likelihood AL calculated one time before corresponding to the sign according to the sign. Input as log likelihood A 1 and add these log likelihoods DGA 01 and A 1. Adder 2 4 6 2 supplies data AM 1 indicating the I monument + I ⁇ obtained by adding the correction term calculation circuits 2 4 7 and selector 2 4 8.
- the correction term calculation circuit 2 4 7 is an adder 2 4 6! And de one data AM 0 supplied from inputs the data AM 1 supplied from the adder 2 4 6 2 calculates the data DM indicating the value of the correction term.
- the correction term calculation circuit 247 uses two differentiators 2 5 1 1, 2 5 1 2 and the value of the correction term in 10 g—sum correction as a table.
- Selective control signal generation for generating control signals for controlling the selection operation by two lookup tables 2 5 2 1 ⁇ 2 5 2 2 and three selectors 2 4 8, 2 5 4, 2 5 5 to be stored It has a circuit 25 3 and two selectors 25 4 and 25 5.
- Differentiator 2 5 1 t is an adder 2 4 6, de Isseki AM 0 supplied from taking the difference between the data AM 1 supplied from the adder 2 4 6 2. Strictly speaking, assuming that each of the data AM 0 and AM 1 consists of, for example, 12 bits, the differentiator 2 51 i has the data of the lower 6 bits of the data AM 0. The difference between the one with "1" added to the most significant bit and the one with "0" added to the most significant bit of the lower 6 bits of data of AM1 is calculated. Differentiator 2 5 1! Supplies the calculated difference value DA 1 to the lookup table 25 2 and the selection control signal generation circuit 25 3.
- the differentiator 2 5 1 de a Isseki AM 1, take the difference between the data AM 0. Strictly speaking, Assuming that the data AMO and AM1 each consist of, for example, 12 bits, the differentiator 251 adds “1” to the most significant bit of the lower 6 bits of the data AMI. The difference between the data with "0" and the data with "0” added to the most significant bit of the lower 6 bits of data of AM0 is calculated. Differentiator 2 5 1 2 supplies a difference value DA 0 calculated in the look-up table 2 5 2 2 and selection control signal generating circuit 2 53.
- the look-up table 2 52 i reads from the table the value of the correction term corresponding to the value of the difference value DA 1 supplied from the difference device 25 1, and supplies it to the selector 254 as data RDA 1. Also, the look-up table 2 52 2 reads the value of the correction term that corresponds to the value of the difference value DA 0 supplied from differentiator 25 1 2 from Te one table, supplied as data RDA 0 to the selector 2 5 4 I do.
- the selection control signal generation circuit 253 generates a control signal SEL for controlling the selection operation by the selectors 248 and 254 based on the data AM0 and AM1 and the difference values DA1 and DA0. And a control signal SL for controlling the selection operation by the selector 255.
- the selection control signal generation circuit 253 like the selection control signal generation circuit 232 described above, determines the upper bit and lower bit of the metric based on the data AM0 and AMI. Are divided to generate control signals SEL and SL indicating a decision sentence for selection, which will be described later.
- the selector 2 54 Based on the control signal SEL supplied from the selection control signal generation circuit 25 3, the selector 2 54 generates the look-up table 2 52! Either the data RDA 1 supplied from the data processor or the data RDA 0 supplied from the look-up table 252 2 is selected. Specifically, if the value of data AM0 is larger than the value of data AM1, the selector 254 determines that the look-up table 2 52 2! Select data from RDA 1. That is, the selector 254 selects the value of the correction term corresponding to the absolute value of the difference between the data AM0 and the data AM1. The selector 254 supplies the selected data CA to the selector 255.
- the selector 255 has data CA supplied from the selector 254 and a predetermined value M based on the control signal SL supplied from the selection control signal generation circuit 253. One of these data. Specifically, since the value of the correction term for the difference value supplied as the data CA has the property of asymptotically approaching a predetermined value, the selector 255 determines that the value of the data CA is If the value exceeds the predetermined value M, a data item having the predetermined value M is selected. The selector 255 supplies the data DM obtained by the selection to the adder 249.
- Such a correction term calculation circuit 247 calculates the value of the correction term in the 10 g-sum correction. At this time, the correction term calculation circuit 247 calculates the values of a plurality of correction terms instead of calculating the absolute value of the difference value between the two pieces of input data and then calculating the value of the correction term, as described later. And select the appropriate one. Further, the correction term calculating circuits 247, the data to be supplied to the differentiator 25 1, the data AM 1 supplied from the adder 246 1 data AM 0 and adder 246 2 is supplied from, but it it For example, assuming that the data consists of 12 bits, the most significant bit of the lower 6 bits of data AM0 is appended with "1", and the lower 6 bits of data AM1 are data.
- the correction term computation circuit 247 the data to be supplied to the differentiator 25 1 2 to that denoted by "0" in the most significant bit Bok of Isseki data of lower six bi Uz DOO data AM 0
- "1" is added to the most significant bit of the lower 6 bits of the data AM1. That is, the differentiator 25 1 252 2, the data in the most significant bit Bok “1” or “0” is attached is supplied in the lower bits of the data supplied from the adder 2461, 246 2 However, this is to perform high-speed comparison of the data AM0 and AM1.
- the selection control signal generation circuit 253 separates the upper bit and the lower bit of the metric. It is related to creating a judgment sentence for selection. This will be described later.
- the selector 248 selects a smaller one of the data AMO and AMI based on the control signal SEL supplied from the selection control signal generation circuit 253.
- the selector 248 supplies the selected data SAM to the adder 249.
- the adder 249 adds the data SAM supplied from the selector 248 and the data DM supplied from the correction term calculation circuit 247 to calculate the log likelihood I ⁇ .
- the adder 247 uses the calculated log likelihood I as the log likelihood CM, Supply 0.
- the I-normalization circuit 250 performs normalization for correcting the bias of the distribution of the log likelihood CM supplied from the adder 249. Various methods are conceivable for this normalization processing, which will be described later.
- the I-normalization circuit 250 also performs a termination process using the termination information TALD.
- the I-normalization circuit 250 performs clipping of the log likelihood I after the normalization according to a required dynamic range and obtains a predetermined 10 g—sum arithmetic circuit 245 245 2 as a log likelihood AL00. , 245 3 , ⁇ • •, 245! Supply to 6 .
- the log likelihood AL 00 after 1 time amount of delay is made by Regis evening not shown, predetermined 1 Og- s um arithmetic circuit 2451, 24 5 2, 245 3, ⁇ ⁇ ⁇ , 245 16 Supplied.
- Such a 10 g-sum arithmetic circuit 245 t calculates and outputs the log likelihood AL00, and combines the data AM0 and AM1 to output the data AG00. That is, 1 og-sum arithmetic circuit 245!
- the predetermined logarithmic likelihood AL00 is used to calculate the logarithmic likelihood I at the next time, so that a predetermined 10 g-sum arithmetic circuit 2451, 245 2 , 245 3 , ⁇ , 245 ! and supplies to the beta, and outputs log-likelihood I human log likelihood I sum I alpha + I ⁇ a indicates to data AG 00 with ⁇ found in log-likelihood I Hino calculation process.
- the 10 g—sum arithmetic circuit 245 2 has the same configuration as the 1 og—sum arithmetic circuit 245 t, and therefore detailed description is omitted, but the log likelihood DGA of the log likelihood DGA 0 2 , DGA 03 and the log likelihood AL calculated one time before, corresponding to the sign, are input as log likelihoods A 0 and A 1, and these log likelihoods DG A 02 and DGA 03 , with a 0, a 1, calculates log likelihood I monument, as a log likelihood AL 0 1, given 1 Og- s um arithmetic circuit 245 245 2, 245 3, ⁇ ⁇ •, 245 16 In addition to this, data AGO 1 indicating the sum I + Ia of the log likelihood I and the one log likelihood Ia is output.
- the 10 g-sum operation circuit 245 3 has the same configuration as the 1 og-sum operation circuit 245, and thus detailed description is omitted, but the log likelihood DGA04 of the log likelihood DGA is omitted.
- DGA 05 and the log likelihood AL calculated one time ago corresponding to the code are input as log likelihoods A 0 and A 1, and these log likelihoods are input.
- the log likelihood I is calculated, and a predetermined 10 g—sum arithmetic circuit 245 245 2 , 245 3,. , 24 5 16 and outputs data AG 02 indicating the sum of the log-likelihood I and the log-likelihood Ia, ie, I + 1 + 1.
- the 10 g-sum operation circuit 245 16 also has the same configuration as the 1 og-sum operation circuit 245, detailed description is omitted, but the log likelihood DGA 30 of the log likelihood DG ⁇ is omitted.
- DGA3 1 and the log likelihood AL calculated one time ago corresponding to the sign are input as log likelihoods A 0 and A 1, and these log likelihoods DGA30, DGA 31, Using A 0 and A l, a log likelihood I is calculated, and a predetermined logarithmic likelihood AL 15 is set as a predetermined 10 g—sum arithmetic circuit 245 ⁇ , 245 2 , 245 3, ⁇ , 245 16 And outputs the data AG 15 indicating the sum I a + I a of the log likelihood I and the log likelihood I a.
- Such an addition / comparison / selection circuit 241 calculates the log likelihood I in a code such that two paths arrive from each state on the trellis to the state at the next time. As will be described later, the addition / comparison / selection circuit 241 does not output the calculated log likelihood I, but outputs the sum I + I of the log likelihood I ⁇ and the log likelihood Ia. That is, the ACS circuit 24 1, 10 g- s um arithmetic circuit 245 245 2, 245 3, ⁇ ⁇ ⁇ , 245 16 it thereby the obtained de Isseki AG 00, AG 0 1, AG 02, ⁇ ⁇ ⁇ ⁇ , AG 15 are bundled and supplied to the selector 244 as an AGT.
- the addition / comparison / selection circuit 242 performs the addition / comparison / selection processing and 10 g processing on a code such that four paths or, depending on the code, eight paths arrive from each state on the trellis to the state at the next time — Perform 10 g — s um operation by adding a correction term using s um correction.
- the addition / comparison / selection circuit 242 uses a code such that four or, depending on the code, eight paths arrive from each state on the trellis to the state at the next time. among them, having a number of 10 g- s um arithmetic circuit 2 56 n of the maximum value among the number of states code to be subjected to decoding.
- the addition / comparison / selection circuit 242 decodes a code having a maximum of eight states. og-sum operation circuit 256..., 256 8 .
- Each 10 g- s um arithmetic circuit 2 56., ⁇ ⁇ ⁇ , distribution of the log likelihood AL for 25 6 8 differs depending on the configuration of the code, here based on the control signal PST, determined by the selector (not shown) or the like Is done.
- the distribution of the log likelihood AL will be described later.
- 10 g- s um arithmetic circuit 2 56 i is five adders 2 57 2 5 72 2 57 3, 2 57 4, 27 1, the value of the correction term in 1 Og- sum correction
- the adder 257 inputs the log likelihood DGA 00 of the log likelihood DGA and, among the log likelihoods AL calculated one time earlier, applies the log likelihood AL corresponding to the sign to the log likelihood.
- the degree A 0 is input, and these log likelihoods DGA 00 and A 0 are added.
- Adder 2 57 t is the log-likelihood I human log likelihood I ⁇ de one evening AM0 showing the sum of the correction term calculation circuit 2 58 l5 2 58 3 obtained by adding, 2 58 5 and selectors Supply to 2 59.
- the adder 257 2 receives the log likelihood DGA 0 1 of the log likelihood DGA, and outputs the log likelihood AL calculated one time before corresponding to the sign according to the sign. Input as log likelihood A 1 and add these log likelihoods DGA 01 and A 1. Adder 2 57 2 supplies data AM 1 indicating the I monument + I ⁇ obtained by adding the correction term calculation circuits 2 58 2 58 4, 2 58 6 and selector 2 59.
- the adder 2 57 3 inputs the log likelihood DGA 0 2 of the log likelihood DGA, and logs the log likelihood AL calculated one time earlier corresponding to the sign according to the sign. Input as likelihood A 2 and add these log likelihoods DGA 0 2 and A 2.
- Adder 2 57 3, supplies the de Isseki AM 2 showing the I alpha + I ⁇ obtained by adding the correction term calculation circuits 2 5 8 2, 2 58 3, 2 58 4 and a selector 2 60 I do.
- the adder 257 4 inputs the log likelihood DGA 03 of the log likelihood DGA and, among the log likelihoods AL calculated one time before, applies the log likelihood AL corresponding to the sign to the log likelihood. Input as degree A 3 and add these log likelihoods DGA 0 3 and A 3.
- the adder 257 4 supplies the data AM 3 indicating the sum of I and I obtained by the addition to the correction term calculation circuit 258 2 , 258 s, 258 ⁇ and the selector 260. I do.
- the correction term calculation circuit 2 58 i has the same configuration as the correction term calculation circuit 247 shown in FIG. 39, the details are omitted here.
- De supplied from - and evening AM0 inputs the data AM 1 supplied from the adder 2 57 2 calculates the data DM 0 indicating the value of the correction term.
- the correction term calculation circuit 2 58! In the same way as the correction term calculation circuit 247, instead of calculating the absolute value of the difference value between the two input data and then calculating the value of the correction term, it calculates the values of a plurality of correction terms. Choose the appropriate one from.
- correction term computation circuit 2 58 the adder 2 57!, 2 57 lower bi Uz Bok topmost Bidzuto of the data AM 0, AM 1 supplied from the 2 "1" or "0"
- the difference between the attached data is calculated and the magnitude comparison of data AM0 and AM1 is performed at high speed.
- Correction term calculation circuit 2 58! Supplies the calculated data DM0 to the selector 268.
- the correction term calculation circuit 258! Generates a control signal SEL0 for controlling the selection operation by the selectors 259, 261, 262, 263, and 264.
- Correction term calculation circuit 2 58 since the same configuration as the correction term calculation circuit 247 shown in FIG. 39 earlier, here it is not described in detail, De supplied from the adder 2 57 3 Isseki AM 2 and the data AM 3 supplied from the adder 2 57 4 Data DM 1 indicating the value is calculated.
- the correction term computation circuit 2 58 like the correction term calculation detection circuit 247, instead of obtaining the value of the correction term from the calculated absolute value of the difference value of the two data input, a plurality of Calculate the value of the correction term and select an appropriate one from them.
- correction term computation circuit 2 58 2 adders 2 57 3, 2 57 4 De supplied from Isseki AM 2, "1" to the most significant bit of the lower Bidzuto of AM 3 or "0 The difference between the data marked with "is calculated and the data AM2 and AM3 are compared at high speed.
- Correction term calculation circuit 2 58 2 supplies the calculated data DM 1 to selector evening 2 68.
- the correction term calculation circuit 258 3 Since the correction term calculation circuit 258 3 has the same configuration as the correction term calculation circuit 247 shown in FIG. 39, the details are omitted here, but the data supplied from the adder 257 0, inputs the data AM2 supplied from the adder 2 5 7 3 calculates the de Isseki DM 2 indicating the value of the correction term. At this time, similarly to the correction term calculation circuit 247, the correction term calculation circuit 258 3 does not calculate the absolute value of the difference between the two pieces of input data and then obtains the value of the correction term. Calculate the value of the correction term and select an appropriate one from them.
- the correction term calculation circuit 2 58 3 the "1" or "0" in the top-bi Uz preparative lower bits of the data AM 0, AM 2 supplied from the adder 2 571, 2 57 3 The difference between the attached data is obtained, and the magnitude comparison of AMO and AM2 is performed at high speed.
- the correction term calculation circuit 258 3 supplies the calculated data DM 2 to the selector 263.
- De correction term calculation circuit 2 58 since the same configuration as the correction term calculation circuit 247 shown in FIG. 39 earlier, here is not described in detail, supplied from the adder 2 57 2 - evening AM 1 and inputs the data AM 2 supplied from the adder 2 57 3, and calculates the data DM 3 indicating the value of the correction term.
- the correction term computation circuit 2 58 like the correction term calculation detection circuit 247, instead of obtaining the value of the correction term from the calculated absolute value of the difference value of the two data input, a plurality of Calculate the value of the correction term and select the appropriate Choose one.
- correction term calculation circuit 2 58 4 In the correction term calculation circuit 2 58 4, adder 2 572 2 5 7 3 at the top bi Uz Bok lower bits of the data AM 1, AM 2 supplied from the "1" or "0 The difference between the data marked with "" is taken, and the data AM I and AM2 are compared at high speed. Correction term calculation circuit 2 58 4 supplies the calculated data DM3 to selector evening 2 63. The correction term computation circuit 2 58 4 finally selector 2 67 2 68 generates a control signal SEL 3 to the control signal SEL 8 for controlling the selecting operation of the selector 2 6 the control signal SEL 3 1 and the control signal generation circuit 270 for selection.
- Correction term calculation circuit 2 58 since the same configuration as the correction term calculation circuit 247 shown in FIG. 39 previously is here not described in detail, the adder 2 57! A de Isseki AM 0 supplied from inputs the data AM 3 supplied from the adder 2 57 4, calculates the data DM4 indicating the value of the correction term.
- the correction term calculating circuit 2 58 similarly to the correction term calculation output circuit 247, rather than from calculates the absolute value of the difference value of 2 Tsunode Isseki entered determine the value of the correction term, Calculate the values of multiple correction terms and select an appropriate one from them.
- correction term computation circuit 2 58 5 with "1" or "0" to the most significant bit of the lower bits of the adder 2 571, 2 57 4 data AM 0 supplied from, AM 3 The difference between the obtained data is taken, and the magnitude comparison of the data AM 0 and AM 3 is performed at high speed.
- Correction term calculation circuit 2 5 8 5 supplies the calculated data DM 4 to selector evening 2 64. Further, the correction term calculating circuit 2 5 8 5 finally generates the control signal SEL as a control signal SEL 8 for controlling the selecting operation of the selector 2 67, 2 68, the control signal SE L 4 Selector 26 2 and the control signal generation circuit 270 for selection.
- De correction term calculation circuit 2 58 6 since the same configuration as the correction term calculation circuit 247 shown in FIG. 39 earlier, here is not described in detail, supplied from the adder 2 5 7 2 - data and AM 1, inputs the data AM 3 supplied from the adder 2 57 4, to calculate the de-Isseki DM 5 indicating the value of the correction term.
- the correction term computation circuit 2 58 6 like the correction term calculation detection circuit 247, instead of obtaining the value of the correction term from the calculated absolute value of the difference value of the two data input, a plurality of Calculate the value of the correction term and select an appropriate one from them.
- correction term calculation circuit 2 58 ⁇ adders 2 57 2 and 2 57 2 57 The difference between the data in which the most significant bit of the lower bits of the data AM 1 and AM 3 supplied from 4 is “1” or “0” is taken, and the data AM 1 Perform high-speed comparison of AM3 and AM3.
- Correction term calculation circuit 2 58 6 supplies the calculated data DM 5 to selector 2 64.
- the correction term computation circuit 2 58 6, finally selector 2 67 generates a control signal SEL 5 to be a control signal SEL 8 for controlling the selecting operation of 2 68, the control signal SEL 5 Selector 2 62 And a selection control signal generation circuit 270.
- Selector 259 is a correction term calculation circuit 2 58! Based on the control signal SEL0 supplied from the controller, the data AM0 or AMI having the smaller value is selected. The selector 259 supplies the data S AM0 obtained by the selection to the selector 267.
- the selector 2 60 based on the control signal SEL 1 supplied from the correction term computation circuit 2 58 2, of the de Isseki AM2, AM3, to select one value is small.
- the selector 260 supplies the data S SAM 1 obtained by the selection to the selector 2 67.
- the selector 261 selects one of the control signals SEL2 and SEL3 based on the control signal SEL0 supplied from the correction term calculation circuit 258. Specifically, when the value of data AM0 is larger than that of data AM1, selector 2661 selects control signal SEL3. The selector 261 supplies the selected control signal SEL6 to the selector 265.
- the selector 262 selects one of the control signals SEL4 and SEL5 based on the control signal SEL0 supplied from the correction term calculation circuit 258. Specifically, the selector 262 selects the control signal SEL5 when the data AM0 is larger than the data AM1. The selector 262 supplies the selected control signal SEL7 to the selector 265.
- the selector 263 selects one of the data DM2 and DM3 based on the control signal SEL0 supplied from the correction term calculation circuit 258. Specifically, the selector 263 selects the data DM3 when the data AM0 has a larger value than the data AM1. The selector 263 supplies the data DS0 obtained by the selection to the selector 266.
- the selector 264 receives the control signal SEL 0 supplied from the correction term calculation circuit 258, Based on this, either DM4 or DM5 is selected. Specifically, the selector 264 selects DM5 when the data AM0 has a larger value than the data AMI. The selector 264 supplies the selected data DS 1 to the selector 266.
- the selector 265 selects one of the control signals SEL 6 and SEL 7 based on the control signal SEL 1 supplied from the correction term calculation circuit 258 2 . Specifically, the selector 265 selects the control signal SEL7 when the value of the data AM2 is larger than the value of the data AM3. The selector 265 supplies the control signal SEL 8 obtained by selection as a control signal for selection in the selectors 267 and 268.
- Selector 2 6 6 based on the control signal SEL 1 supplied from the correction term computation circuit 2 58 2, data DSO, among DS 1, selects either. More specifically, the selector 266 selects DS 1 if the data AM 2 has a larger value than the data AM 3. The selector 266 supplies the selected data DS 2 to the selector 269.
- the selector 267 selects one of the data SAM0 and SAM1 based on the control signal SEL8. Specifically, when the control signal SEL8 is the control signal SEL7, the selector 267 selects the data SAM1. The selector 267 supplies the selected data SAM 2 to the adder 27 1. The selector 268 selects one of the data DM0 and DM1 based on the control signal SEL8. Specifically, when the control signal SEL8 is the control signal SEL7, the selector 268 selects the data DM1 overnight. The selector 268 supplies the selected data DS 3 to the selector 269.
- the selector 269 selects one of the data DS2 and DS3 based on the control signal SEL9 supplied from the selection control signal generation circuit 270.
- the selector 269 supplies the selected data RDM to the adder 271.
- the selection control signal generation circuit 270 generates a control signal SEL 9 for controlling the selection operation by the selector 269 based on the control signals SEL 2, SEL 3, SEL 4, and SEL 5.
- the selection control signal generation circuit 270 By taking the logical sum of the logical product of SEL 2, SEL 3, SEL 4, and SEL 5 and the negation of the logical product of the control signals SEL 2, SEL 3, SEL 4, and SEL 5, the control signal SEL 9 is obtained.
- the adder 271 adds the data SAM2 supplied from the selector 267 and the data RDM supplied from the selector 269 to calculate a log likelihood I.
- the adder 271 supplies the calculated log likelihood I as a log likelihood CM to the I-normalization circuit 272.
- the I-normalization circuit 272 performs the normalization for correcting the bias of the distribution of the log likelihood CM supplied from the adder 271, similarly to the I ⁇ normalization circuit 250 described above.
- the I-normalization circuit 272 also performs termination processing using the termination information TALD.
- the I-normalization circuit 272 performs a clipping of the log-likelihood I after the normalization according to a required dynamic range, and obtains a predetermined 10 g—sum operation circuit 2 56 as a log-likelihood AL00. - - -, and supplies it to the 2 5 6 8.
- the log likelihood AL 00 after 1 time amount of delay is made by Regis evening not shown, predetermined log- s um arithmetic circuit 2 5 6 - - -, it is supplied to the 2 56 8.
- Such a 10 g-sum arithmetic circuit 25 6, calculates and outputs the log likelihood AL 00, and bundles the data AMO, AM I, AM 2, and AM 3 and outputs the data as AG 00. . That is, the log-sum operation circuit 25 6 i uses the determined log-likelihood AL00 to calculate the log-likelihood I at the next time. ⁇ ⁇ ⁇ supplies 2 5 6 8, and outputs the data AG 00 indicating the sum I monument + I ⁇ the logarithmic likelihood I human log likelihood I ⁇ obtained in the process of calculating the log likelihood I monument .
- 10 g-sum arithmetic circuit 2 56! Is the data AM0, AM I, AM 2, A ⁇ ⁇ indicating the likelihood corresponding to the four paths that have reached each state, or the four paths obtained by bundling eight paths depending on the code.
- the data AMO, AMI, AM2 and the data corresponding to two or more paths are obtained, and the data corresponding to the path with the highest likelihood, the most likely path, is selected from the data corresponding to these paths.
- the 1 og—sum arithmetic circuit 2 56 t calculates the data AM0, AM I, AM 2, and AM 3 by performing an operation analogous to a so-called victory battle, so that the value of the data AM 0 and the data AM I
- the value of the data AM2 and the value of the data AM3 are compared, and the data corresponding to the maximum likelihood path is selected.
- the 10 g-sum arithmetic circuit 2 5 6 8 is a 1 og-sum arithmetic circuit 2 56!
- the log likelihood DGA 28, DGA 29, DGA 30, DGA 31 of the log likelihood DGA and the logarithm calculated one time before Among the likelihoods AL those corresponding to the codes are input as log likelihoods A 0, A 1, A 2, and A 3, and these log likelihoods DGA 28, DGA 29, DGA 30, DGA 31, AO, A with 1, a 2, A3, calculates a log likelihood I monument, as a log likelihood AL 07, a predetermined 10 g- s um arithmetic circuit 2 5 6 - - - supplies 2 5 6 8 Then, data AG 07 indicating the sum I «+ ⁇ of the log likelihood I and the log likelihood I is output.
- Such an addition / comparison / selection circuit 242 calculates the log likelihood I in a code such that four or, depending on the code, eight paths arrive from each state on the trellis to the state at the next time. I do. Like the above-described addition / comparison / selection circuit 241, the addition / comparison / selection circuit 242 does not output the calculated log likelihood I, but adds the log likelihood I to the one log likelihood I y 1 «+ Output 1 key. That is, the ACS selection ⁇ path 242 bundled 10 g- s um arithmetic circuit 2 5 6 - - - 2 5 6 de Isseki AGO 0 it thus obtained 8, ⁇ ⁇ ⁇ , the AG 07, The data is supplied to the selector 244 as AGF.
- the ACS circuit 242 log- sum operation circuit 2 56 - - -, 2 5 6 8 log likelihood AL 00 it thereby obtained of,..., Bundling AL 07, the log likelihood AL Is supplied to the I + 1 + 1 calculation circuit 243.
- the addition / comparison / selection circuit 242 is originally provided to obtain the log likelihood I in a code such that four paths reach each state on the trellis to the state at the next time.
- the log likelihood I ⁇ of a code that can reach eight paths can be obtained depending on the code. This is described in detail in "5-5-3" and "5-5_5".
- the I + I key calculation circuit 243 for example, It is provided for decoding a code having a parallel path on the trellis, such as a code by a convolutional encoder, and calculates the sum of log likelihood I and log likelihood Ia.
- the I ⁇ + I key calculation circuit 243 includes three selectors 273, 274, and 275, and here, four I key + I key calculation cell circuits. 27 6 27 6 2 , 276 3 , and 276 4 .
- the selector 273 selects one of the predetermined log likelihoods AL 0 0 and AL 01 corresponding to the code, from among the log likelihoods AL supplied from the addition / comparison / selection circuit 242. Select one of them.
- the selector 273 supplies the log likelihood AL 01 S obtained by the selection to the I + I key calculation cell circuits 276 276 2 , 276 3, and 276 4 .
- the selector 274 selects one of the predetermined log likelihoods AL 0 1 and AL 0 2 of the log likelihood AL supplied from the addition / comparison / selection circuit 242 corresponding to the code. Select one of them.
- the selector 274 supplies the log likelihood AL 02 S obtained by selecting I «+ I ⁇ calculating cell circuits 27 6 2 76 2 2 7 6 3, 276 4.
- the selector 275 based on the number-of-memory information MN, selects one of the predetermined log likelihoods AL 01 and AL 03 corresponding to the code, of the log likelihoods AL supplied from the addition / comparison / selection circuit 242. Select one of them.
- the selector 27 5 supplies the log likelihood AL 03 S obtained by selecting I monument + I ⁇ calculating cell circuit 2 7 6 2 76 2 2 7 6 3, 276 4.
- I Hi + I key calculation cell circuit 27 6! Has eight adders 2 77 ,, 2 77 2, 2 77 3, 277 4, 27 7 5, 2 77 ⁇ , 2 777, 2 77 8.
- Adder 277 Is a predetermined log likelihood D GAB 00 corresponding to the sign of the log likelihood D GAB supplied from the Ia distribution circuit 157, and a log likelihood AL supplied from the addition / comparison / selection circuit 242.
- the predetermined log likelihood AL 00 corresponding to the code is added.
- the adder 277i outputs the data obtained by the addition as data AM0.
- the adder 277 2 includes a predetermined log likelihood D GAB 0 1 corresponding to the sign of the log likelihood D GAB supplied from the I distribution circuit 157 and an addition comparison selection circuit 242. Of the log likelihoods AL supplied from the corresponding log likelihood AL 00 corresponding to the sign. The adder 277 2 and outputs the Isseki de obtained by adding the data AM 1.
- Adder 2 77 3 outputs a Isseki de obtained by the addition as data AM2.
- Adder 27 7 4 outputs a Isseki de obtained by the addition as data AM3.
- the adder 277 5 includes a predetermined log likelihood D GAB 04 corresponding to the code of the log likelihood D GAB supplied from the I distribution circuit 157, and a log likelihood supplied from the selector 274. The degree AL 0 2 S is added. Adder 2 77 5 outputs a Isseki de obtained by the addition as data AM4.
- the adder 2 e 7 e is provided with a predetermined log likelihood D GAB 05 corresponding to the sign of the log likelihood D GAB supplied from the I distribution circuit 157 and a log supplied from the selector 274. Add the likelihood AL 02 S. Adder 2 7 7 6, De is obtained by adding - outputs the data as the data AM 5.
- the adder 277 7 includes a predetermined log likelihood D GAB 06 corresponding to the code of the log likelihood D GAB supplied from the I distribution circuit 157, and a log likelihood supplied from the selector 275. Add the degree AL 03 S. Adder 2 7 7 7 outputs Isseki de obtained by the addition as data AM6.
- the adder 277 8 includes a predetermined log likelihood D GAB 07 corresponding to the code of the log likelihood D GAB supplied from the I distribution circuit 157, and a log likelihood supplied from the selector 275. Add the degree AL 03 S. Adder 2 7 7 8 outputs Isseki de obtained by the addition as data AM7.
- a log likelihood D GAB indicating the log likelihood Ia obtained by the I ⁇ ⁇ ⁇ distribution circuit 157 in a state where the parallel paths are not bundled
- the log likelihood AL calculated by the arithmetic comparison selection circuit 242 is added to the sum of the log likelihood I and the log likelihood Ia used to obtain the log soft output I when the parallel paths are bundled.
- I ⁇ + I key calculation cell circuit 27 6! Outputs the calculated data AMO, AMI, AM2, AM3, ⁇ 4, AM5, AM6, AM7 as data AG00.
- I monument + I ⁇ calculating cell circuit 2 7 6 since the same configuration as the I monument + I ⁇ calculating cell circuit 27 6, the detailed description is omitted, among the log likelihood DGAB, code
- I monument + 1 ⁇ calculating cell circuit 2 7 6 2 outputs the calculated data as data AG O 1.
- I monument + I ⁇ calculating cell circuit 276 since the same configuration as the I monument + I ⁇ calculating cell circuit 2 7 6, the detailed description is omitted, among the log likelihood DGAB, the reference numeral The corresponding log likelihood D GAB 16, DGAB 17, DGAB 18, DGAB 19, DGAB 20, DGAB 21, DGAB 22, D GAB 23, and the sign of the log likelihood AL
- the parallel logarithm is bundled using the predetermined log likelihood AL 00 and the log likelihood AL OIS, AL 02 S and AL 03 S corresponding to The sum of log likelihood I and log likelihood I y is calculated.
- I monument + I ⁇ calculating cell circuit 2 7 6 3 outputs the calculated data as data AG 02.
- the I «+ I key calculation cell circuit 276 4 has the same configuration as the I key + I key calculation cell circuit 276, detailed description is omitted, but the log likelihood DGAB That is, the predetermined log likelihood D GAB 24, DGAB 25, DGA B 26, DGAB 27, DGAB 28, DGAB 29, DGAB 30, DGAB 31 corresponding to the code, and the log likelihood AL
- the parallel paths are bundled using the predetermined log likelihood AL 00 corresponding to the code and the log likelihood ALOIS, AL 02 S, AL 03 S. In this case, the sum of the log likelihood I and the log likelihood I a used in obtaining the log soft output I person is calculated.
- I monument + I ⁇ calculating cell circuit 2 7 6 4 outputs the calculated data as data AG 0 3.
- Such an I + Ia calculation circuit 243 calculates the sum of the log likelihood I and the log likelihood Ia, and calculates the calculated data AG00, AG01, AG02, AG03. And supply them to selectors 2 4 4 as AGE.
- the selector 244 based on the input bit number information IN, compares the data AG which indicates the sum of the log likelihood I and the log likelihood I supplied from the addition / comparison / selection circuit 241 with an addition comparison.
- the data AGF indicating the sum of the log likelihood I and the log likelihood Ia supplied from the selection circuit 24 and the log likelihood I and the logarithmic likelihood supplied from the I + 1 calculation circuit 24 Select any one of the data AGE, which indicates the sum with Ia.
- the selector 244 determines that the code by the element encoder in the encoder 1 has no parallel path on the trellis and that two paths arrive from each state to the state at the next time.
- the code by the element encoder in the encoding device 1 is changed from each state to the state at the next time when there is no parallel path on the trellis. If the code is such that the number of paths reaches, the data AGF is selected, and the code by the element encoder in the encoding device 1 is represented by a trellis having at most 32 branches, and If the code has a maximum of 4 states, more specifically, if there are parallel paths on the trellis such that 8 paths reach each state for 4 states, Select the data A GE. That is, here, the input bit number information IN is used as a control signal for controlling the selection operation by the selector 244, but in practice, the control signal indicated by the code configuration is input to the selector 244. Is performed.
- the I-parameter calculation circuit 158 calculates the log-likelihood I and does not output the calculated log-likelihood ⁇ as it is, but instead calculates the log-likelihood I used for calculating the log soft output I input.
- the sum with the log-likelihood Ia is output as data AG.
- This data AG is supplied to the soft output calculation circuit 161 as data AGD after a predetermined delay.
- the I /? Calculation circuit 159 calculates the log likelihood I? Using the log likelihoods DGB0 and DGB1 supplied from the I distribution circuit 157. Specifically, based on the notation described at the beginning of “2.”, the 1/7 calculation circuit 1559 uses the log likelihood I
- the calculation shown in (52) is performed, and the log likelihood I? Of the two systems at each time t is calculated in parallel.
- the operator “#” in the following equation (52) indicates the 10 g-sum operation as described above, and transitions from state m to state m at input “0”. It shows a 10 g-sum operation of the log likelihood at the time and the log likelihood at the time of transition from state m, 'to state m at input "1". More specifically, if the constant s gn is “+1”, the I? Calculation circuit 159 performs the operation shown in the following equation (53), and on the other hand, the constant s gn becomes “1 1”. In the case of, the log likelihood I?
- the I ⁇ calculation circuit 159 calculates, for each received value yt, the probability /? Calculate the log likelihood I with the sign of degree I? Or probability?
- the I? Calculation circuit 159 includes the generation matrix information CG supplied from the control circuit 60, the input bit number information IN supplied from the code information generation circuit 151, the type information WM, and the memory.
- the calculation circuit 159 supplies the calculated log likelihood I ⁇ of the two systems to the I3 storage circuit 160 as log likelihood ⁇ 0, ⁇ 1.
- the 1? Calculation circuit 159 includes a control signal generation circuit 280 that generates a control signal, and one log likelihood I 3 of two systems of log likelihood I 3.
- a control signal generation circuit 280 that generates a control signal
- the control signal generation circuit 280 uses the generation matrix information CG, the input bit number information IN, the type information WM, and the memory number information MN to generate four lines from each state on the trellis to a state at the next time.
- the state of the transition destination in the code that reaches the path is calculated and supplied to the addition / comparison / selection circuit 281 for I-0 and the addition / comparison / selection circuit 282 for I31 as a control signal NST.
- the I ⁇ 0 addition / comparison / selection circuit 281 is provided for calculating the log likelihood I ⁇ 0.
- the I? 0 addition / comparison / selection circuit 2811 performs an addition / comparison / selection process and a 10-g Addition and comparison selection circuit 283 that performs processing to add a correction term by sum correction, and four paths from each state on the trellis to the state at the next time, or eight paths depending on the sign For codes such as An addition comparison selection circuit 284 that performs a comparison selection process and a process of adding a correction term by 10 g-sum correction, and a selector 285 that performs a two-to-one selection.
- the addition / comparison / selection circuit 283 adds a correction term by addition / comparison / selection processing and 10-sum correction to a code such that two paths arrive from each state on the trellis to the state at the next time. 10 g—sum operation is performed.
- the addition / comparison / selection circuit 283 includes two paths from each state on the trellis to the state at the next time, similarly to the addition / comparison / selection circuit 241 described above.
- the addition / comparison / selection circuit 283 decodes a code having a maximum of 16 states, and 16 1 og-sum operation circuits 286 286 2 , 286 3 ,. 86 1 ⁇ .
- log-sum operation circuits 2861, 286 2 , 286 3 , ⁇ , 286 16 are based on the transitions on the trellis, and the log likelihood I of the branch corresponding to the output pattern on the trellis I And the log-likelihood I? 0 one time ago in each state are supplied. That is, 10 g—sum arithmetic circuit 286 286 2 , 286 3 , ⁇ , 286 16 has the log likelihood of the branch corresponding to the output pattern on the trellis in log likelihood DGB 0 respectively. The one corresponding to the Ia and the one corresponding to the log likelihood I ⁇ 0 in each state among the calculated log likelihood BTT one time ago are supplied.
- the log-sum operation circuits 286 !, 286 2 , 286 3) ..., 286 16 each obtain the log likelihood I? In each state at the next time as the log likelihood BTT.
- Each 10 g—sum arithmetic circuit 286! , 286 2 , 286 3 , ... , 286 ie the distribution of the log likelihood BTT differs according to the code configuration, and is determined here by a selector (not shown) based on the memory number information MN. The distribution of the log-likelihood BTT will be further described later.
- 10 g-sum arithmetic circuit 286! Are the three adders 287 !, 2872, 290, the correction term calculation circuit 288 for calculating the value of the correction term in the 10 g-sum correction, the selector 289, and 1? 0 normalization circuit 29 1.
- the adder 287 t inputs the log likelihood DGB 00 of the log likelihood DGB 0, and the log likelihood B TT calculated one time earlier corresponding to the sign according to the code. Input as the degree B 0 and add these log likelihoods D GB 00 and B 0.
- Adder 2 87! Supplies the data AM 0 indicating the sum of the log likelihood I ⁇ and the log likelihood Ir obtained by the addition to the correction term calculation circuit 288 and the selector 289.
- the adder 2 87 2 inputs the log likelihood DGB 0 1 of the log likelihood DGB 0 and logs the log likelihood BTT calculated one time earlier corresponding to the sign according to the sign Input as the likelihood B 1 and add these log likelihoods D GB 0 1 and B 1.
- Adder 2 87 2 is supplied to the correction Ko ⁇ detection circuit 2 88, and a selector 289 data AM 1 indicating the I? 0 + I ⁇ obtained by adding.
- the correction term calculation circuit 288 Since the correction term calculation circuit 288 has the same configuration as the correction term calculation circuit 247 shown in FIG. 39, the details are omitted here. De supplied from - and evening AM0, inputs the de Isseki AM 1 supplied from the adder 2 87 2 calculates the data DM indicating the value of the correction term. At this time, similarly to the correction term calculation circuit 247, the correction term calculation circuit 288 calculates a plurality of correction terms instead of calculating the absolute value of the difference value between the two input data and then obtaining the value of the correction term. Calculate the value of and select the appropriate one from them.
- the correction term computation circuit 2 88 with "1" or "0" to the most significant bits of the lower-bi Uz Bok of the data AM 0, AM I supplied from the adder 2 871, 2 87 2 The difference between the obtained data is taken and the magnitude comparison of the data AM0 and AMI is performed at high speed.
- the correction term calculation circuit 288 supplies the calculated data DM to the adder 290. Further, the correction term calculation circuit 288 generates a control signal SEL for controlling the selection operation by the selector 289.
- the selector 289 selects one of AMO and AMI having a smaller value based on the control signal SEL supplied from the correction term calculation circuit 288.
- the selector 289 supplies the data S SAM obtained by the selection to the adder 290.
- the adder 290 adds the data SAM supplied from the selector 289 and the data DM supplied from the correction term calculation circuit 288 to calculate the log likelihood I ⁇ 0.
- the adder 290 supplies the calculated log likelihood I ⁇ 0 to the I ⁇ 0 normalization circuit 291 as log likelihood CM.
- the I ⁇ 0 normalization circuit 291 performs the normalization for correcting the bias of the distribution of the log likelihood CM supplied from the adder 290, similarly to the I normalization circuit 250 described above.
- the I-0 normalization circuit 291 also performs termination processing using termination information TB0D.
- the I ⁇ 0 normalization circuit 29 1 performs the clipping of the normalized log likelihood I?
- a predetermined logarithmic likelihood ⁇ ⁇ 00 as a predetermined 10 g-sum arithmetic circuit. 2 86 2 86 2, 2 8 6 3, ⁇ ⁇ ⁇ , and supplies the 2 8 6 16.
- the log likelihood BT 00 after 1 time amount of delay is made by a not-shown register, given 10 g- s um arithmetic circuit 28 61, 2 86 2, 28 6 3, ⁇ ⁇ ⁇ , 2 8 Supplied to 6 1 ⁇ .
- Such a log-sum operation circuit 286 i obtains and outputs the log likelihood BTOO. That is, the log-sum operation circuit 28 6! Is used to calculate the log likelihood I? 0 at the next time by using a predetermined 10 g-sum arithmetic circuit 2 8 6 286 2 , 286 3 , ⁇ , 2 8 6 16 and output to the outside.
- log-sum operation circuit 286 2 Since the log-sum operation circuit 286 2 has the same configuration as the 1 og-sum operation circuit 286, detailed description is omitted, but the log likelihood DGB 02 of the log likelihood DGB 0 is omitted.
- DGB 03 and the log likelihood ⁇ ⁇ 1 calculated one time earlier that correspond to the code are input as log likelihood ⁇ 0, B 1, and these log likelihoods DGB 02, D Using GB 03, BO, ⁇ 1, log likelihood I 0 is calculated, and a predetermined 10 g—sum arithmetic circuit 286 2 86 2 , 286 3 , ⁇ , is calculated as log likelihood BTO 1. 28 6! 6 and output to outside.
- 10 g- s um arithmetic circuit 2 8 6 3 for the same configuration as 1 Og- sum operation circuit 2 8 6 i, the detailed description is omitted, the logarithm of the log likelihood DGB 0 Likelihood DGB 04, DGB 05, and the log likelihood BTT calculated one time before are input as log likelihood B 0, B 1 according to the sign, and these log likelihoods are input.
- the log likelihood I? 0 is calculated, and the log likelihood BT 0 2 is given as a predetermined 10 g-sum arithmetic circuit 286! , 2 8 6 2 2 2 86 3, ⁇ ⁇ ⁇ , and supplies to 286 16, and outputs to the outside.
- the 10 g—sum arithmetic circuit 28 6 16 is also a 1 og—sum arithmetic circuit 28 6 Although the detailed description is omitted because it has the same configuration as, the log likelihood DGB 30 and DGB 31 of the log likelihood DGB 0 and the log likelihood BTT calculated one time earlier are Input the log likelihood B0, B1 corresponding to the code, and calculate the log likelihood I? 0 using these log likelihoods DGB30, DGB31, B0, B1. and, as a log likelihood BT 1 5, given 1 Og- s um arithmetic circuit 28 6 ,, 28 6 2 2 8 6 3, ⁇ ⁇ ⁇ , and supplies to 286 ie, output to the outside.
- Such an addition / comparison / selection circuit 283 calculates the log likelihood I-0 of a code such that two paths reach from each state on the trellis to the state at the next time.
- the addition / comparison / selection circuit 283 is composed of 1 og-sum arithmetic circuit 2 8 6 28 6 286,..., And 2 816 16 , data BT 00, BT 0 1, BT 02,--. , BT 15 are bundled and supplied to the selector 285 as log likelihood BTT.
- the addition / comparison / selection circuit 284 performs an addition / comparison / selection process for a code such that four or, depending on the code, eight paths arrive from each state on the trellis to the state at the next time. Perform 10 g-sum calculation by adding a correction term by 1 og-sum correction.
- the addition / comparison / selection circuit 284 has four lines from each state on the trellis to the state at the next time, as shown in FIG. 44, or
- the code has a 1 og-sum operation circuit 292 n which is the maximum value of the number of states of the code to be decoded among the codes that eight paths can reach depending on the code.
- ACS circuit 284, and performs decoding of a code having a maximum of 8 stearyl Ichito shall have eight 10 g- s um arithmetic circuit 2 92 ⁇ ⁇ •, 29 2 8 .
- the one corresponding to the log likelihood Ia of the branch corresponding to the force pattern and the one corresponding to the log likelihood ITF0 in each state of the calculated log likelihood BTF one time ago are supplied.
- the distribution of the log-likelihood BTF for each 10 g—sum arithmetic circuit 2 921, ⁇ , 292 8 depends on the code configuration.
- a selector (not shown) is used. It is determined.
- the distribution of the log-likelihood BTF will be further described later.
- 10 g- s um arithmetic circuit 292 t includes five adders 293 2 9 3 2 29 3 3 29 3 4, 307, exits ⁇ the value of the correction term in 1 Og- sum correction 6 correction term calculation circuits 294 2 942, 294 3 , 294 4 , 294 5 , 294 6 , and 1 selector 2 295, 296, 297, 298, 299, 300, 301, 302, 303, 304, 305, a selection control signal generation circuit 306 for generating a control signal for controlling the selection operation by the selector 305, and an I-0 normalization circuit 308.
- the adder 293i inputs the log likelihood D GB 00 of the log likelihood D GB 0 and simultaneously, among the log likelihood BTFs calculated one time ago, the one corresponding to the sign. Input as log likelihood B 0 and add these log likelihoods DGB 00, B 0.
- the adder 293 i generates a correction term calculation circuit 294 2 94 3 , 2 94 5 which calculates the data AM 0 indicating the sum of the log likelihood I ⁇ 0 and the log likelihood Ia obtained by the addition. And selector 295.
- the adder 293 2 receives the log likelihood DGB 0 1 of the log likelihood D GB 0 and inputs the log likelihood B TF calculated one time earlier according to the sign. Is input as the log likelihood B 1 and these log likelihoods D GB 0 1 and B 1 are added. Adder 2 93 2 is supplied to the circuit 2 94 2 94 4 294 6 and the selector 29 5 out correction term calculation data AM 1 indicating the I / 50 + I ⁇ obtained by adding.
- the adder 293 3 are both by entering the logarithmic likelihood D GB 0 2 of the log likelihood D GB 0, among the log likelihood BTF computed one time before, also of the relevant in accordance with the code Input as log likelihood B 2 and add these log likelihoods D GB 02, B 2.
- the adder 2933 supplies the data AM 2 indicating I ⁇ 0 + I key obtained by the addition to the correction term calculating circuits 294 2 , 294 3 , 294 4 and the selector 296.
- the adder 293 inputs the log likelihood DGB 03 of the log likelihood DGB 0 and, among the log likelihood BTFs calculated one time ago, the log likelihood BTF corresponding to the code. Input as B3 and add these log likelihoods DGB03, B3.
- Adder 2 9 3 4 supplies the de Isseki AM 3 showing the I 50 + I ⁇ obtained by adding the correction term calculation detection circuit 294 2, 2 94 5, 294 beta and selector 29 6.
- the correction term calculation circuit 294 Since the correction term calculation circuit 294 has the same configuration as the correction term calculation circuit 247 shown in FIG. 39, its details are omitted here.
- the correction term calculation circuit 294 When inputs the data AM 1 supplied from the adder 2 9 3 2 calculates the data DM0 indicating the value of the correction term.
- the correction term calculation circuit 294 similarly to the correction term calculation circuit 247, the correction term calculation circuit 294 does not calculate the absolute value of the difference between the two pieces of input data and then obtains the value of the correction term. Calculate the value of the correction term and select an appropriate one from them. Also, the correction term calculation circuit 2 94!
- 2 93 most significant bits of the lower-bi Uz bets of Isseki de supplied AM 0, AM 1 from 2 "1" or "0" between attached data Take the difference and compare the data AM0 and AM1 at high speed.
- the correction term calculation circuit 294 supplies the calculated data DM0 to the selector 304. Further, the correction term calculation circuit 294! Generates a control signal SEL0 for controlling the selection operation by the selectors 295, 297, 298, 299, 300.
- the correction term calculation circuit 294 2 Since the correction term calculation circuit 294 2 has the same configuration as the correction term calculation circuit 247 shown in FIG. 39, the details are omitted here, but the data supplied from the adder 293 3 2, inputs the data AM 3 supplied from the adder 2 93 4, calculates the data DM 1 indicating the value of the correction term. In this case, the correction term computation circuit 294 2 is corrected similarly to the term calculated detection circuit 247, instead of obtaining the value of the correction term from the calculated absolute value of the difference value of the two data input, a plurality of correction Calculate the value of the term, and select an appropriate one from them.
- correction term computation circuit 2 94 2 the adder 2 9 33, 2 93 4 De supplied from Isseki AM 2, "1" to the most significant bit of the lower bits of the AM 3 or " The difference between the data marked with "0" is calculated and the data AM2 and AM3 Perform small comparisons at high speed.
- Correction term calculation circuit 2 94 2 supplies the calculated de Isseki the DM 1 to selector evening 304.
- the correction term calculation circuit 294 3 Since the correction term calculation circuit 294 3 has the same configuration as the correction term calculation circuit 247 shown in FIG. 39 earlier, the details are omitted here, but the data supplied from the adder 293 And the data AM2 supplied from the adder 293 3 are input, and data DM2 indicating the value of the correction term is calculated.
- the correction term computation circuit 2 94 like the correction term calculation detection circuit 247, instead of obtaining the value of the correction term from the calculated absolute value of the difference value of the two data input, a plurality of Calculate the value of the correction term and select an appropriate one from them.
- correction term calculation circuit 2 94 3 an adder 2 93 ,, 2 93 3 De supplied from Isseki AM 0, "1" at the top Bidzuto backward bi Uz bets of AM 2 or " The difference between data marked with "0" is calculated, and the data AM0 and AM2 are compared at high speed.
- Correction term calculation circuit 2 94 3 supplies the calculated de Isseki DM 2 to selector evening 2 9 9.
- the correction term computation circuit 294 3 is finally selector 303: 304 to generate a control signal SEL 2 to be controlled signal SEL 8 for controlling the selecting operation of, the control signal SEL 2 selector 2 9 7 and It is supplied to the selection control signal generation circuit 306.
- De correction term calculation circuit 2 94 since the same configuration as the correction term calculation circuit 247 shown in FIG. 39 earlier, here is not described in detail, supplied from the adder 2 93 2 - evening AM type and I, and a de-Isseki AM2 supplied from the adder 293 3, to calculate the de-Isseki DM3 indicating the value of the correction term.
- the correction term computation circuit 2 94 like the correction term calculation output circuit 247, rather than from calculates the absolute value of the difference value of 2 Tsunode Isseki entered determine the value of the correction term, Calculate the values of multiple correction terms and select an appropriate one from them.
- correction term calculation circuit 2 94 4 the adder 2 9 3 2, 2 93 most significant bits of the lower-bi Uz bets of Isseki de supplied AM 1, AM 2 from 3 "1" or The difference between the data marked with "0" is calculated, and the magnitude comparison of AM I and AM 2 is performed at high speed.
- Correction term calculation circuit 2 94 4 supplies the calculated data DM3 to selector 2 9 9. Further, the correction term calculation circuit 294 4 finally generates a control signal SEL 8 which is a control signal SEL 8 for controlling the selection operation by the selectors 303 and 304. 3 and supplies this control signal SEL 3 to the selector 297 and the selection control signal generation circuit 306.
- De correction term calculation circuit 2 94 5 since the same configuration as the correction term calculation circuit 247 shown in FIG. 39 earlier, here is omitted details are supplied from the adder 2 93 i - evening AM0
- the correction term calculating circuit 2 94 similarly to the correction term calculation output circuit 247, rather than from calculates the absolute value of the difference value of 2 Tsunode Isseki entered determine the value of the correction term, Calculate the values of multiple correction terms and select an appropriate one from them.
- the correction term computation circuit 2 94 5 an adder 2 931, 2 93 4 at the top bi Uz Bok backward bi Uz bets of the data AM 0, AM 3 supplied from
- Correction term calculation circuit 2 94 5 supplies the calculated data DM 4 to selector evening 300.
- the correction term computation circuit 2 94 5 finally Selected evening 303: 304 to generate a control signal SEL 4 as a control signal SEL 8 for controlling the selecting operation of, the control signal SEL 4 selector 2 9 8 and the control signal generation circuit 306 for selection.
- De correction term calculation circuit 2 94 6 since the same configuration as the correction term calculation circuit 247 shown in FIG. 39 earlier, here is not described in detail, supplied from the adder 2 93 2 - motor AM 1 and the data AM 3 supplied from the adder 293 4 are input, and data DM 5 indicating the value of the correction term is calculated.
- the correction term computation circuit 2 94 6 like the correction term calculation output circuit 247, rather than from calculates the absolute value of the difference value of 2 Tsunode Isseki entered determine the value of the correction term, Calculate the values of multiple correction terms and select an appropriate one from them.
- the correction term computation circuit 294 beta is an adder 2 9 3 2> 2 9 3 4 Isseki de supplied AM 1, the top-bi Uz Bok lower bits of the AM 3 from
- Correction term calculation circuit 2 94 6 supplies the calculated data DM 5 to selector evening 300. Further, the correction term calculation circuit 2946 generates a control signal SEL 5 which is a control signal SEL 8 for finally controlling the selection operation by the selectors 303 and 304, and this control signal SEL 5 is used as the selector 298 and the selection signal SEL 5. Control signal generation times Supply to Road 306.
- the selector 295 is a correction term calculation circuit 2 94! Based on the control signal SEL0 supplied from the controller, the data AM0 or AMI having the smaller value is selected. The selector 295 supplies the data SAM 0 obtained by the selection to the selector 303.
- the selector 29 6, based on the control signal SEL 1 supplied from the correction term computation circuit 2 94 2, of the data AM2, AM3, to select one value is small.
- the selector 296 supplies the selected data SAM1 to the selector 303.
- the selector 297 selects one of the control signals SEL2 and SEL3 based on the control signal SEL0 supplied from the correction term calculation circuit 294. Specifically, the selector 297 selects the control signal SEL3 when the value of the data AM0 is larger than that of the data AM1. The selector 297 supplies the selected control signal SEL6 to the selector 301.
- the selector 298 selects one of the control signals SEL4 and SEL5 based on the control signal SEL0 supplied from the correction term calculation circuit 294. Specifically, the selector 298 selects the control signal SEL5 when the data AM0 is larger than the data AM1. The selector 298 supplies the selected control signal SEL 7 to the selector 301.
- the selector 299 is a correction term calculation circuit 2 94! One of DM2 and DM3 is selected based on the control signal SEL0 supplied from the controller. Specifically, the selector 299 selects the data DM 3 when the value of the data AM 0 is larger than the data AM 1. The selector 299 supplies the selected data DS0 to the selector 302.
- the selector 300 selects one of the data DM4 and DM5 based on the control signal SEL0 supplied from the correction term calculation circuit 294i. Specifically, the selector 300 selects the data DM5 when the data AM0 has a larger value than the data AM1. The selector 300 supplies the selected data DS1 to the selector 302.
- the selector 30 based on the control signal SEL 1 supplied from the correction term computation circuit 2 94 2, of the control signal SEL 6, SE L 7, selects either.
- the selector 301 selects the control signal SEL7 when the value of the data AM2 is larger than the value of the data AM3.
- the selector 301 supplies the control signal SEL 8 obtained by selection as a control signal for selection in the selectors 303 and 304.
- the selector 302 based on the control signal SEL 1 supplied from the correction term computation circuit 2 94 2, of the Isseki de DS 0, DS 1, selects either. Specifically, the selector 302 selects the data DS 1 when the data AM 2 is larger than the data AM 3. The selector 302 supplies the selected data DS2 to the selector 305.
- Selector 303 selects one of data SAM0 and SAM1 based on control signal SEL8. Specifically, when the control signal SEL8 is the control signal SEL7, the selector 303 selects the data SAM1. The selector 303 supplies the data S SAM 2 obtained by the selection to the adder 307. Selector 304 selects one of data DM 0 and DM 1 based on control signal SEL 8. Specifically, when the control signal SEL8 is the control signal SEL7, the selector 304 selects the data DM1. The selector 304 supplies the selected data DS3 to the selector 305.
- the selector 305 selects one of the data DS2 and DS3 based on the control signal SEL9 supplied from the selection control signal generation circuit 306.
- the selector 305 supplies the selected data RDM to the adder 307.
- the selection control signal generation circuit 306 generates a control signal SEL9 for controlling the selection operation by the selector 305 based on the control signals SEL2, SEL3, SEL4, and SEL5. Specifically, the selection control signal generation circuit 306 calculates the logical product of the control signals SEL 2, SEL 3, SEL 4, and SEL 5 and the logical product of the control signals SEL 2, SEL 3, SEL 4, and SEL 5.
- the control signal SEL 9 is generated by taking the logical OR with the negation.
- the adder 307 adds the data SAM 2 supplied from the selector 303 and the data RDM supplied from the selector 305 to calculate a log likelihood I ⁇ 0.
- the adder 307 uses the calculated log likelihood I ⁇ 0 as the log likelihood CM and an I / 50 normalization circuit. Supply 308.
- the I-0 normalization circuit 308 performs the normalization for correcting the bias of the distribution of the log likelihood CM supplied from the adder 307, similarly to the I-0 normalization circuit 291 described above.
- the I-0 normalization circuit 308 also performs termination processing using the termination information TB0D.
- the I 50 normalization circuit 308 performs clipping on the log likelihood I? 0 after the normalization according to a required dynamic range, and as a log likelihood BT 00, a predetermined 10 g—sum arithmetic circuit 2921, - - -, and supplies to 292 8. In this case, log likelihood BT 00, after 1 time amount of delay is made by Regis evening not shown, a predetermined 10 g- s um arithmetic circuit 292 - - -, are supplied to 292 8.
- Such a log-sum operation circuit 292! Calculates and outputs the log likelihood BT00. That is, the log-sum operation circuit 292! Is the log likelihood BT 00 determined, for use in calculating the log likelihood I 0 at the next time, a predetermined 10 g- sum Starring ⁇ path 202!, ⁇ ⁇ ⁇ , And supplies to 292 8, the external Output to
- the 10 g-sum operation circuit 292 i calculates the likelihood corresponding to the four paths that have reached each state or four sets obtained by bundling eight paths depending on the code. Evening By comparing the magnitude of likelihood for all combinations of data corresponding to two paths selected from AM0, AMI, AM2, and AM3, these data AMO, AMI, and AM2 are compared. , ⁇ 3, the data corresponding to at least two or more paths with the highest likelihood is obtained, and the data corresponding to the most likely path, which is the path with the highest likelihood, is obtained from the data corresponding to these paths.
- the 10 g-sum arithmetic circuit 292 by performing an operation analogous to a so-called win battle on the data AM 0, AM I, AM 2, and AM 3, obtains the value of the data AM 0 and the data AM Compare the value of I, the value of AM2 and the value of data AM3, and select the data corresponding to the maximum likelihood path.
- the 10 g-sum operation circuit 292 8 has the same configuration as the 1 og-sum operation circuit 292 i, and therefore detailed description is omitted, but the log likelihood DGB 28 of the log likelihood DGB 0 is omitted.
- DGB 29, DGB 30, DGB 31 1, and among the log likelihood BTFs calculated one time ago, those corresponding to the code are input as log likelihoods B 0, B 1, B 2, B 3
- these log likelihoods D GB 28, DGB 29, DGB 3 0, D GB 31, BO, B 1, B 2, B 3 are used to calculate the log likelihood I? 0, and as a log likelihood BT 07, a predetermined log-sum operation circuit 29 2 ⁇ ⁇ supplies 2 9 2 8, and outputs to the outside.
- Such an addition / comparison / selection circuit 284 calculates the log likelihood I ⁇ 0 in a code such that four paths, or depending on the code, eight paths reach from each state on the trellis to the state at the next time. calculate. ACS circuit 284, 1 Og- s um arithmetic circuit 2 9 2 - - - 2 92 8 it thereby de obtained Isseki BT 00,..., Bundled BT 0 7, as a data BTF Supply to selector 285.
- the addition / comparison / selection circuit 284 like the above-described addition / comparison / selection circuit 242, is originally a log likelihood I in a code such that four paths arrive from each state on the trellis to the state at the next time. Although it is provided to obtain 0, as described above, it is possible to obtain the log likelihood I 0 in a code such that eight paths reach depending on the code. This is described in detail in "5-5-3" and "5-5-5".
- the selector 285 supplies, based on the input bit number information IN, the log likelihood ⁇ ⁇ ⁇ ⁇ indicating the log likelihood I ⁇ 0 supplied from the addition / comparison / selection circuit 283 and the supply from the addition / comparison / selection circuit 284 Either one of the data BTF indicating the log likelihood 1 to 0 is selected. Specifically, the selector 285 determines that the code by the element encoder in the encoder 1 has two paths from each state to the state at the next time when there is no parallel path on the trellis.
- the log likelihood ⁇ ⁇ ⁇ is selected, and the code by the element coder in the coding device 1 is set so that no parallel path exists on the trellis and each state If the code is such that four paths arrive from the state to the state at the next time, the log likelihood BTF is selected. That is, here, the input bit number information I I is used as a control signal for controlling the selection operation by the selector 285, but in reality, the control signal indicated by the code configuration is input to the selector 285. .
- the addition / comparison / selection circuit for I-0 like this calculates the log likelihood I-0 and outputs the calculated log likelihood I; 50 as log likelihood ⁇ 0.
- This log likelihood ⁇ 0 is supplied to the I / 5 storage circuit 160.
- the I-1 addition / comparison / selection circuit 282 is provided for calculating the log likelihood I-1. Since the I / 51 addition / comparison / selection circuit 282 has the same configuration as the 1-0 addition / comparison / selection circuit 281, detailed description is omitted, but the log likelihood D GB 0 and the termination information TB 0 D Instead, log likelihood D GB 1 and termination information TB 1 D are input to calculate log likelihood I ⁇ 1, and the calculated log likelihood I ⁇ 1 is output as log likelihood B 1.
- This log likelihood B 1 is supplied to an I ⁇ storage circuit 160 ( such an I? Calculation circuit 159 calculates two log likelihoods I? 0 and I? 1 in parallel, The calculated log likelihood I? 0, 1/51 is calculated as
- the memory circuit 160 includes, for example, a plurality of banks of RAMs, a control circuit, and a selection circuit.
- the I5 storage circuit 160 stores the log likelihoods B0 and B1 supplied from the I? Calculation circuit 159.
- the I? Storage circuit 160 selects predetermined information from the stored log likelihoods BO and B1 by a selection circuit under the control of an internal control circuit, and outputs a logarithmic soft output I person.
- the logarithmic likelihood BT used for the calculation is supplied to the soft output calculation circuit 16 1.
- the element decoder 50 is described in International Publication No.W 099/62 183 as a method of memory management in the I? Storage circuit 160 when performing sliding window processing.
- the logarithmic softness is finally achieved.
- the output I ⁇ can be obtained in the original time-series order.
- the soft output calculation circuit 16 1 is composed of the data AG D supplied from the I calculation circuit 1 58,
- the calculation shown in (55) is performed to calculate the log soft output I ⁇ at each time t.
- the operator “# ⁇ ” in the following equation (55) indicates a cumulative addition operation of the 10 g-sum operation represented by the above-described operator “#”. Further, the soft output calculating circuit 161 can also calculate the logarithmic soft output I ⁇ in units of symbols or bits.
- the soft-output calculation circuit 16 1 outputs the externally supplied output data selection control signal CI ⁇ , the prior probability information format information CAP ⁇ supplied from the control circuit 60, and the code information generation circuit 15 1 Based on the supplied input bit number information IN, the memory number information MN, and the branch input / output information BI0, the logarithmic soft output I person corresponding to the information symbol or the posterior probability information for the information bit, or the code symbol or Calculates the log soft output I person corresponding to the posterior probability information for the code bit.
- the soft output power calculation circuit 16 1 is an external information calculation circuit that uses the log soft output I calculated in symbol units or the log soft output I calculated in bit units as log soft outputs SLM and BLM, respectively. 16 3, amplitude adjustment and clipping circuit 16 4, and hard decision circuit 16 5.
- the soft output calculation circuit 16 1 calculates I + I a to calculate the sum of the log likelihood Ia, the log likelihood Ia, and the log likelihood I.
- + I? Calculation circuit 3 1 enable signal generation circuit 3 1 1 for generating enable signal, and, for example, 6 10 g—sum operation circuits 3 1 2 3 1 2 2 , 3 1 2 3 , 3 1 2 4 , 3 1 2 5 , 3 1 2 6, and an I-input calculation circuit 3 13 for calculating the log soft output I input.
- the I + I + I ⁇ calculation circuit 310 is composed of an I? Distribution circuit 314 for distributing the log likelihood I? And a maximum value of the number of states of the code to be decoded, here 3 2 adders 3 1 5!, 3 1 5 2) 3 1 5 3 , 3 1 5 4 , 3 1 5 5 , 3 1 5 6 , 3 1 53 1, 3 1 532 .
- the I? Distribution circuit 314 distributes the log likelihood BT supplied from the I? Storage circuit 160 according to the code configuration. That is, the 1/3 distribution circuit 3 1 4 depends on the code configuration The log likelihood BT is distributed so as to correspond to the trellis. At this time, the 1? Distribution circuit 314 distributes the log likelihood BT based on the input bit number information IN supplied from the code information generation circuit 151. The distribution circuit 3 1 4 adds the log likelihood I 5 obtained by the distribution to the adder 3 1 5 3 1 5 2 , 3 1 5 3 , 3 1 54, 3 1 5 5 , 3 1 5 e, ⁇ ⁇ ⁇ ⁇ , 3 1 53 1, 3 1 5 32 That is, the I? Distribution circuit 3 14 uses the log likelihood I5 used for calculation of the log soft output I?
- the adder 315 is composed of the data AG 0 0 of the data A GD indicating the sum of the log likelihood I and the log likelihood Ia supplied from the I
- the log likelihood B TD 00 of the log likelihood B TD supplied from 3 14 is added.
- the adder 315 i outputs the log likelihood I and the sum of the log likelihood Ia and the log likelihood I ⁇ h obtained as the addition as data AGB 00.
- Adder 3 1 5 2 the de Isseki AG O 1 of the data AGD supplied from I ratio calculating circuit 1 5 8, of log likelihood BTD supplied from the I? Distribution circuit 3 1 4 Log likelihood BTD 0 0 is added.
- the adder 3 1 5 2 outputs the log likelihood I 1 and the sum of the log likelihood I y and the log likelihood I 5 obtained by the addition as AGB 01.
- the adder 3 1 5 3 is composed of the data AG 0 2 of the data AGD supplied from the I calculation circuit 15 8 and the pair of the log likelihood BTD supplied from the I distribution circuit 3 14. Add the number likelihood BTD 01.
- Adder 3 1 5 3 outputs the sum of the log-likelihood I monument and logarithmic likelihood I ⁇ a logarithmic likelihood 1?
- the adder 3 1 5 4 outputs the data A G O 3 of the data A GD supplied from the I calculation circuit 15 8 and the log likelihood BTD supplied from the I 2 distribution circuit 3 14.
- Log likelihood BTD 0 1 is added.
- Adder 3 1 5 4 outputs the sum of the log-likelihood I monument and logarithmic likelihood I ⁇ a log likelihood I? Obtained by the addition as data AGB 0 3.
- the adder 3 1 5 5 is composed of the data AGD 4 supplied from the I calculation circuit 15 8 and the data AGD 4 and the log likelihood BTD supplied from the I distribution circuit 3 14. Log likelihood BTD 0 2 is added.
- the adder 3 1 5 5 outputs the sum of the log likelihood Ia, the log likelihood Ia, and the log likelihood 1/5 obtained by the addition as data AGB04.
- the adder 315 includes a data AG0 5 of the data AGD supplied from the I-height calculation circuit 158 and a pair of the log likelihood BTD supplied from the I ⁇ distribution circuit 314. The number likelihood BTD 0 2 is added.
- Adder 3 1 5 6 outputs the sum of the log-likelihood I human log likelihood I I and log likelihood 1? Obtained by the addition de Isseki as AGB 0 5.
- Adder 3 1 5 31 includes a data AG 3 0 of the de Isseki AGD supplied from I alpha calculator 1 5 8, of log likelihood BTD supplied from I ⁇ distributing circuit 3 1 4 Add log likelihood B TD 15.
- Adder 3 1 5 31 outputs the sum of the summed log likelihood I human log likelihood obtained I ⁇ the logarithmic likelihood I? As data AGB 3 0.
- Adder 3 1 5 32 includes a data AG 3 1 of the data AGD supplied from I ratio calculating circuit 1 5 8, I? Log-likelihood of the log likelihood BTD supplied from the distribution circuit 3 1 4 The degree B TD 15 is added.
- the adder 3 1 5 32 outputs the sum of the log likelihood I 1 log likelihood I ⁇ obtained by the addition and the log likelihood 15 as data AGB 31.
- the I-I + Ia + I? Calculation circuit 310 calculates the sum of the log likelihood I, the log likelihood Ia and the log likelihood I, and calculates the calculated data AGB 0 0, AG ⁇ 0 1, AGB 0 2, AGB 03, AGB 04, AGB 05, ⁇ , AGB 30, AGB 31, and a log-sum operation circuit 3 1 2!, 3 1 as data AGB 2 2 , 3 1
- the enable signal generation circuit 3 1 1 is connected to the selector 3 2 3 3 2 3 2 , 3 2 3 3 , 3 2
- a selection control signal generation circuit 316 for generating a control signal for controlling the selection operation by 3 4 , a symbol selection branch selection circuit 3 19, and a bit selection branch selection circuit 3 2 0, 3 2 1, 3
- the selection control signal generation circuit 3 16 converts the output data selection control signal CI ⁇ ⁇ ⁇ supplied from the outside and the prior probability information format information CAP ⁇ supplied from the control circuit 60 into Based on, it generates a control signal AP for controlling the selecting operation of the selector 32 3 32 3 2 j 323 3 , 323 4.
- the effective branch selection circuit 3 17 is based on the input bit number information IN and the memory number information MN supplied from the code information generation circuit 15 1, and the symbol corresponding branch selection circuit 3 19 and the bit corresponding branch selection circuit are provided. 3 20, 32 1, and 32 2 generate control signals M 1, M 2, and M 3 indicating whether or not the branch input / output information BI 0 input thereto is valid. That is, the effective branch selection circuit 317 is used to select the branch to be selected by each of the symbol corresponding branch selection circuit 319 and the bit corresponding branch selection circuit 320, 321, 322. To generate the control signals M 1, M 2, and M 3.
- the effective branch selection circuit 3 17 supplies the generated control signals M 1 and M 2 to the bit corresponding branch selection circuits 32 0, 32 1 and 322, and supplies the control signal M 3 to the symbol corresponding branch selection circuit 31 9 and bits are supplied to the corresponding branch selection circuit 32 0, 32 1, 32 2.
- the output data selection circuit 3 18 outputs the code information generation circuit 1 based on the output data selection control signal CITM supplied from the outside and the input bit number information IN supplied from the code information generation circuit 15 1. 5 From the branch input / output information BI 0 supplied from 1, the one corresponding to the branch corresponding to the code configuration is selected. The output data selection circuit 318 supplies the selected branch input / output information BI00 to the bit corresponding branch selection circuit 320 and supplies the selected branch input / output information BI01 to the bit corresponding branch selection circuit 321. At the same time, the selected branch input / output information BI02 is supplied to the corresponding bit selection circuit 3222.
- the symbol-corresponding branch selection circuit 319 is provided for calculating the log soft output I ⁇ in symbol units.
- the symbol-corresponding branch selection circuit 319 selects a branch corresponding to the symbol using the branch input / output information ⁇ I 0 supplied from the code information generation circuit 15 1. At this time, the symbol corresponding branch selection circuit 319 selects a branch based on the control signal # 3 supplied from the effective branch selection circuit 317.
- the symbol corresponding branch selection circuit 3 19 generates an enable signal S ⁇ ⁇ 0, S EN 1, SEN 2, S which indicates whether the input corresponding to the selected branch is “0” or “1”.
- the bit corresponding branch selection circuit 320 is provided for calculating the logarithmic soft output I input in bit units.
- the bit corresponding branch selection circuit 320 uses the branch input / output information B 1 ⁇ 0 supplied from the output data selection circuit 318 to select a branch corresponding to the bit.
- the bit corresponding branch selection circuit 320 selects a branch based on the control signals M1, M2, and M3 supplied from the effective branch selection circuit 317.
- the bit corresponding branch selection circuit 320 generates enable signals EN 00 and EN 01 indicating whether the input corresponding to the selected branch is “0” or “1”, and outputs the enable signal EN 00.
- the bit corresponding branch selection circuit 321 is provided to calculate the logarithmic soft output I person in bit units, similarly to the bit corresponding branch selection circuit 320.
- the bit corresponding branch selection circuit 321 selects the branch corresponding to the bit by using the branch input / output information BI01 supplied from the output data selection circuit 318. At this time, the bit corresponding branch selection circuit 321 selects a branch based on the control signals M 1, M 2, and M 3 supplied from the effective branch selection circuit 3 17.
- the bit-corresponding branch selection circuit 32 1 generates enable signals EN 10 and EN 11 indicating whether the input corresponding to the selected branch is “0” or “1”, and generates the enable signal. supplies the EN 1 0 to the selector 323 3 supplies the rice one enable signal EN 1 1 to the selector 323 4.
- the bit corresponding branch selection circuit 322 is provided for calculating the logarithmic soft output I in bit units, similarly to the bit corresponding branch selection circuit 320. Using the branch input / output information BI02 supplied from the output data selection circuit 318, the bit corresponding branch selection circuit 322 selects a branch corresponding to the bit. At this time, the bit corresponding branch selection circuit 322 selects a branch based on the control signals M1, M2, and M3 supplied from the valid branch selection circuit 317. The bit corresponding branch selection circuit 322 generates an enable signal EN20 or EN21 indicating whether the input corresponding to the selected branch is "0" or "1", and outputs the enable signal EN20. supplies to 10 g- s um arithmetic circuit 3 1 2 5, supplying the rice one enable signal EN 2 1 to 10 g- s um arithmetic circuit 3 1 2 6 I do.
- the selector 323 based on the control signal AP supplied from the selection control signal generating circuit 3 1 6, and Ineburu signal S EN 1 supplied from Shinporu corresponding branch selection circuit 3 19, bi Uz preparative corresponding branch selection circuit Either of the enable signal E NO 1 supplied from 320 is selected. Specifically, the selector 323 2, control signals the AP, information Shinporu or information bit output data selection ⁇ control signal CI TM to the effect of outputting information for indicates, and prior probability information indicating a symbol unit If it is indicated by the format information CAPP, the enable signal SEN1 supplied from the symbol corresponding branch selection circuit 319 is selected.
- the selector 323 were selected rice - supplies enable signal ENS 1 to 10 g- s um arithmetic circuit 3 1 2 2.
- the selector 323 based on the control signal AP supplied from the selection control signal generating circuit 3 1 6, and Ineburu signal S EN2 supplied from Shinporu corresponding branch selection circuit 3 1 9, bit corresponding branch selection circuit 32 Select one of the enable signals EN 10 supplied from 1.
- the control signal AP is indicative output data selection ⁇ control signal CI TM to the effect of outputting the information for the information symbols or information bits Bok, and prior probability information indicating that Shinporu unit If the type information CAPP indicates, the enable signal SEN2 supplied from the symbol corresponding branch selection circuit 319 is selected.
- the selector 323 4 controls the control signal A supplied from the selection control signal generation circuit 3 16. Either the enable signal S EN3 supplied from the symbol corresponding branch selection circuit 3 19 or the enable signal EN 11 supplied from the bit corresponding branch selection circuit 32 1 based on P. Select one or the other.
- the selector 3 23 4 control signal AP is, information Shinporu or information bit output data selection ⁇ control signal CI TM to the effect of outputting information for indicates, and, prior probability indicating that Shinporu unit If the information is indicated by the information format information CAPP, the enable signal SEN3 supplied from the symbol corresponding branch selection circuit 319 is selected.
- Such an enable signal generation circuit 311 uses the output data selection control signal CITM, the prior probability information format information CAPP, the number of memories information MN, and the branch input / output information BI0 to generate an image corresponding to the selected branch. Generates the ENS0, ENS1, ENS2, ENS3, EN20, and EN21 signals and generates 1 og—sum operation circuit 3 1 2 i, 3 1 22, 3 1 23, 3 1 2 4 , 3 Supplied to 1 2 5 , 3 1 2 e .
- the log-sum operation circuit 3 1 2! Decodes a code having a maximum of 16 states, and has 31 10 g 1 sum operation cell circuits 325,..., 32531.
- the 10 g—sum operation cell circuit 325 is composed of two differentiators 32 6 3 2 6 2 , six selectors 3 27, 328, 329, 332, 336, 338, and selectors 3 2 7, 32
- a selection control signal generation circuit 330 for generating a control signal for controlling the selection operation by 8, 32 9 and a selection control signal generation circuit 331 for generating a control signal for controlling the selection operation by the selector 332 ,
- the differentiator 326 calculates the difference between the predetermined data A GB 000 and AGB 00 1 corresponding to the sign in the data AG B supplied from the 1 + 1 key + 1? Calculation circuit 310. . Strictly speaking, the differentiator 326 i has the lower order of the data AGB 000, assuming that the data AGB 000 and AGB 00 1 each consist of, for example, 13 bits. The most significant bit of the 6-bit data with "1" appended, and the most significant bit of the lower 6 bits of AGB001 with "0” appended to the most significant bit And take the difference. Differ 32 6! Supplies the calculated difference value DA 1 to the selector 3 27 and the selection control signal generation circuit 330.
- Differentiator 3 26 2 of the data AG B supplied from the I monument + I ⁇ + I / 5 calculation circuit 3 1 0, a difference component of a given data AGB 00 1, AGB 000 applicable in accordance with the code Take. Strictly speaking, the differentiator 32 6 2 data AGB 000, AGB 00 1, but it it, for example 1 3 When made of bits, the data AGB 00 1 of lower six bi Uz bets most significant bit of data The difference between the one with "1” appended to the data and the one with "0” appended to the most significant bit of the lower 6 bits of AGB 000 is calculated. Differentiator 32 6 2 supplies a difference value DA 0 calculated the selector 3 28 and selection control signal producing formation circuit 330.
- the selector 327 generates a differentiator 326! Based on the control signal SL1 supplied from the selection control signal generation circuit 330. One selected from the difference value DA1 supplied from the PC and the data having the predetermined value N1. Specifically, since the value of the correction term for the difference value DA 1 has the property of asymptotically approaching a predetermined value, the selector 327 sets the value of the difference value DA 1 to a predetermined value N 1. If so, select a data item having a predetermined value N1. The selector 327 supplies the selected data S081 to the selector 329.
- the selector 328 based on the control signal SL 1 supplied from the selection control signal generating circuit 330, a difference value DA 0 supplied from the differentiator 3 2 6 2, and Isseki de having a predetermined value N 1 Select one of them. Specifically, since the value of the correction term for the difference value DA 0 has a property of asymptotically approaching a predetermined value, the selector 328 determines that the value of the difference value DA 0 exceeds the predetermined value N 1 If so, select the data having the predetermined value N1. The selector 328 supplies the selected data SDA0 to the selector 329.
- the selector 329 based on the control signal SL2 supplied from the selection control signal generation circuit 330, receives the data SDA1 supplied from the selector 327 and the data SDA0 supplied from the selector 328, Select one of them. Concrete When the value of data AGB 000 is larger than the value of data AGB 00 1, selector 329 selects data SDA 1 supplied from selector 327, and selector 329 selects and obtains the value. The obtained data DM is supplied to the lookup table 335.
- the selection control signal generation circuit 330 generates a control signal SL1 for controlling the selection operation by the selectors 327 and 328 based on the data AGB00 and AGB01 and the difference values DAI and DA0. At the same time, a control signal SL2 for controlling the selection operation by the selector 329 is generated. The selection control signal generation circuit 330 also supplies the generated control signal SL2 to the selection control signal generation circuit 331. At this time, the selection control signal generation circuit 330, like the selection control signal generation circuit 232 described above, uses the upper bits and lower bits of the metric based on the data AGB00 and AGB01. The control bits SL 1 and SL 2 indicating the decision sentence for selection are generated by dividing the upper bits and the lower bits, which will be described later.
- the selection control signal generation circuit 33 1 is based on the enable signals EN 000 and EN 00 1 of the enable signal EN S 0 supplied from the enable signal generation circuit 3 1 1 and the control signal SL 2. Then, a control signal SEL for controlling the selection operation by the selector 332 is generated.
- the selector 332 selects one of the data AGB 000 and AGB 001 based on the control signal SEL supplied from the selection control signal generation circuit 331.
- the selector 332 supplies the selected data DAG to the adder 337.
- the AND gate 333 ANDs the enable signals EN000 and EN001.
- the AND gate 333 supplies the obtained logical product E NA to the selector 336 as a control signal for selection.
- Look-up table 335 stores the value of the correction term in the 10 g-sum correction. —Remember as a bull.
- the look-up table 335 reads the value of the correction term corresponding to the value of the data DM supplied from the selector 329 from the table and supplies it to the selector 336 as data RDM.
- the selector 336 selects one of the data RDM supplied from the look-up table 335 and the data having a predetermined value N2 based on the logical product ENA supplied from the AND gate 333. I do. Specifically, when the logical product ENA is "1", the selector 336 selects the data RDM. The selector 336 supplies the selected data SDM to the adder 337.
- the predetermined value N2 is an offset value added so as to unify the positive / negative identification code of the data CAG described later. That is, the data DAG, which is one of the data AGB 000 and AGB 001, may take a value that crosses both positive and negative values. However, expressing both positive and negative values will increase the circuit scale. Invite. Thus, in the log_sum operation cell circuit 325, a predetermined value N2 to be added by an adder 337 described later is introduced so as to unify the positive / negative identification code of the data DAG.
- the adder 337 adds the data DAG supplied from the selector 332 and the data SDM supplied from the selector 336.
- the adder 337 supplies the calculated data CAG to the selector 338.
- the selector 338 selects one of the data CAG supplied from the adder 337 and the data having a predetermined value N3 based on the logical sum EN supplied from the OR gate 334. . Specifically, when the logical sum EN is “1”, the selector 338 selects the data CAG.
- the selector 338 supplies the data AG L obtained by selecting 10 g- s um operation cell circuit 325 17.
- Such a log-sum operation circuit 325 is composed of the data A GB 000 and the data AGB 00 1 supplied from the I + 1 key + calculation circuit 310 and the enable signal generation circuit 311 Using the supplied enable signal EN 000 and enable signal EN 00 1, as described later, by performing an operation analogous to the first round in the winning game, a logarithmic soft output I person is calculated.
- One 10g-sum operation in the cumulative addition operation of the 10g-sum operation performed at this time is performed.
- the 10 g-sum arithmetic circuit 325 converts the calculated data AGL to AGB 100 Performs an operation analogous to the second round in the winning game. 1 og—sum operation cell circuit 325 17 and enables signal EN 100 at 10 g—sum operation cell circuit 325! Supply 7
- the log-sum arithmetic circuit 325 2 is a 1 og-sum arithmetic circuit 325! Since the configuration is the same as that described above, a detailed description is omitted, but the data AGB 002 and data AGB 003 supplied from the I + I + I5 calculation circuit 310 and the enable signal generation circuit 3 1 Using the enable signal EN 002 and enable signal EN 003 supplied from 1 and performing an operation analogous to the first round in the winning game, the one in the cumulative addition operation of the log-sum operation is performed.
- the 10 g—sum operation circuit 325 2 supplies the calculated data AGL as the data AGB 101 to the log—sum operation cell circuit 325, and outputs the enable signal EN 101 to 1 og— s um supplies the operation cell circuit 325 17.
- the 10 g—sum operation circuit 32 5 3 has the same configuration as the 1 og—sum operation circuit 325 i, a detailed description is omitted.
- the first round in the winning game By performing an operation analogous to the above, one 10 g-sum operation in the cumulative addition operation of the log-sum operation is performed.
- the 1 og—sum arithmetic circuit 325 3 performs the operation analogous to the second round in the winning game using the calculated data AGL as the data AGB 102 10 g—sum arithmetic cell circuit 325! Supplies to 8, supplies Ineburu signal EN 1 0 2 to 10 g- s um operation cell circuit 325 18.
- 1 Og- s um arithmetic circuit 3 2 5 5 stores the calculated data AGL as data AGB 1 04, the 1 0 g- s um operation cell circuit 3 2 5 19 performs an operation to be likened to a second round of the tournament supplies, supplies rice one enable signal EN 1 0 4 to 1 Og- s um operation cell circuit 3 2 5 19.
- 1 0 g—sum operation circuit 3 2 5 7 is 1 og—sum operation circuit 3 2 5! Although the detailed description is omitted because it has the same configuration as that of the data AGB 0 1 2 and the data AGB 0 13 supplied from the I «+ I + I?
- a log-sum operation is performed by performing an operation analogous to the first round in the winning game.
- Performs one 10g-sum operation in the cumulative addition operation of. 1 0 g—sum operation circuit 3 2 5 7 performs the operation analogous to the second round in the winning game by using the calculated data AGL as data AGB 106.
- the enable signal EN106 to 10 g—sum operation cell circuit 325 2 . To supply.
- the log-sum arithmetic circuit 325 8 is a 1 og-sum arithmetic circuit 325! Since the configuration is the same as that described above, the detailed description is omitted, but the data AGB 014 and the data AGB 0 15 supplied from the I + I + I? By using the enable signal EN014 and the enable signal EN015 supplied from the single signal generation circuit 311 1 to perform an operation analogous to the first round in the winning match, the log-sum Performs one 10 g-sum operation in the cumulative addition operation. log- s um arithmetic circuit 32 5 8, the computed data AGL as data AGB 1 07, log- s um operation cell circuit 325 2. The enable signal EN 107 is supplied to the log-sum operation cell circuit 325 2 . To supply.
- the 1 og-sum arithmetic circuit 325 9 has the same configuration as that of the log-sum arithmetic circuit 325 i, and therefore detailed description is omitted, but it is supplied from the I / I + I + I? Using the data AGB 0 16 and the data bus AG B 0 17 and the enable signal EN 0 16 and the enable signal EN 0 17 supplied from the enable signal generation circuit 3 11 By performing an operation that can be compared to the first round in, one 10g-sum operation in the cumulative addition operation of the log-sum operation is performed.
- the 10 g-sum arithmetic circuit 325 9 supplies the calculated data AGL to the 10 g-sum arithmetic cell circuit 325 2 , which performs the operation analogous to the second round in the winning game, as the data AGB 108.
- the enable signal EN 108 is supplied to the 10 g-sum operation cell circuit 325 21 .
- the 10 g—sum arithmetic circuit 325 has the same configuration as the 1 og—sum arithmetic circuit 325>, so a detailed description is omitted, but it is supplied from the I + I + I5 calculation circuit 310.
- the first round of the winning game By performing an operation analogous to, one og-sum operation in the cumulative addition operation of the log-sum operation is performed.
- the 10 g—sum operation circuit 325 u supplies the calculated data AGL as data AGB 110 and performs an operation analogous to the second round in the winning match.
- 10 g—sum operation cell circuit 325 22 together, supply the rice one enable signal EN 1 1 0 to 1 Og- s um operation cell circuit 325 22.
- the 10 g—sum arithmetic circuit 325 12 has the same configuration as the 1 og—sum arithmetic circuit 325, and therefore detailed description is omitted. Using the supplied data AGB 022 and data AGB 023, and the enable signal ⁇ ⁇ 022 and enable signal EN 023 supplied from the enable signal generation circuit 311, the first round in the winning game By performing an operation analogous to the above, one 10 g-sum operation in the cumulative addition operation of the log-sum operation is performed.
- the log-sum operation circuit 325 12 supplies the calculated data AGL as the data AGB 11 1 to the 10 g-sum operation cell circuit 325 22 and supplies the enable signal EN 11 1 It supplied to g- s um operation cell circuit 32 5 22.
- the log-sum arithmetic circuit 325 13 Since the log-sum arithmetic circuit 325 13 has the same configuration as the log-sum arithmetic circuit 325 i, a detailed description is omitted, but it is supplied from the I-I + I-A + I-calculation circuit 310. Using the data AGB 024 and data AGB 025, and the enable signal EN 024 and enable signal EN 025 supplied from the enable signal generation circuit 311 in the first round of the winning game. By performing an operation that can be compared, one 10 g-sum operation in the cumulative addition operation of the log-sum operation is performed. 10 g—sum operation circuit 325 13 decompresses the calculated data AGL As the AGB 1 12, perform the operation analogous to the second round in the winning game. 10 g—Supply the sum operation cell circuit 3 2 5 23 and the enable signal EN 11 2 10 g— It is supplied to the sum operation cell circuit 3 25 23 .
- 10 g-sum arithmetic circuit 3 2 5 14 is 1 og—sum arithmetic circuit 3 2 5! Since the configuration is the same as that described above, the detailed description is omitted, but the data AGB 0 26 and the data AGB 0 27 supplied from the I + I + I? Calculation circuit 3110 and the enable By using the enable signal EN026 and the enable signal EN027 supplied from the signal generation circuit 311 1 to perform an operation corresponding to the first round in the winning game, log-sum Performs one 10g-sum operation in the cumulative addition operation.
- the 1 og—sum arithmetic circuit 3 25 H supplies the calculated data AGL to the log—sum arithmetic cell circuit 3 25 23 as the data AGB 113, and also enables the enable signal EN 1 1 3 Is supplied to the 10 g—sum operation cell circuit 3 25 23 .
- 1 0 g—sum operation circuit 3 2 5 15 is a log—sum operation circuit 3 2 5! Since the configuration is the same as that described above, a detailed description is omitted, but the data AGB 0 28 and the data AGB 0 29, supplied from the I + I + I / 5 calculation circuit 3 10, And using the enable signal EN028 and the enable signal EN029 supplied from the enable signal generation circuit 311 to perform an operation analogous to the first round in a winning game. Performs the one 10 g—sum operation in the cumulative addition operation of the log—sum operation. 10 g—sum arithmetic circuit 3 2 5 15 uses the calculated data AGL as data AGB 114 to perform an operation analogous to the second round in the winning game. 10 g-sum arithmetic cell circuit 3 supplies to the 2 5 24 supplies rice one enable signal EN 1 1 4 to 1 0 g- s um operation cell circuit 3 2 5 24.
- the log-sum operation circuit 3 2 5 16 Since the log-sum operation circuit 3 2 5 16 has the same configuration as the log-sum operation circuit 3 25 t, detailed description is omitted, but I-I + I-A + I? Data AGB 0 3 0 and data AGB 0 3 1 supplied from 0 and an enable signal EN 0 3 0 and an enable signal EN 0 3 1 supplied from the enable signal generation circuit 3 1 1
- the 1 og—sum operation circuit 325 16 supplies the calculated data AGL to the log—sum operation cell circuit 325 24 as the data AGB 115, and also enables the enable signal EN 115. Is supplied to the 10 g—sum operation cell circuit 325 24 .
- the 1 og-sum operation circuit 325 17 has the same configuration as the 1 og-sum operation circuit 325 t, the detailed description is omitted, but the 1 og-sum operation cell circuit 325 17 is supplied from the 1 og-sum operation cell circuit 32 5. Using the data AGB 100 and the enable signal EN 100 and the data AGB 10 1 and the enable signal EN 101 supplied from the 10 g-sum operation cell circuit 325 2. By performing an operation analogous to the second round of the winning match, one log-sum operation in the cumulative addition operation of the 10 g-sum operation is performed.
- the 10 g-sum arithmetic circuit 325 18 is a 1 og-sum arithmetic circuit 325! Since the same configuration as a detailed description is omitted, 1 Og- s um operation cell circuit 32 5 3 Data AG B 102 and rice one enable signal EN 102 supplied from, as well as, 10 g- s um operation cell circuit 325 4 data a GB 103 and rice supplied from - using enable signal EN 103, by performing an operation to be likened to a second round of the tournament, one in log- s um accumulating operation operations Performs a log-sum operation of.
- the 10 g—sum operation circuit 325 18 supplies the calculated data AGL as data AGB 201 to the log—sum operation cell circuit 325 25 , and outputs the enable signal EN 201 1 to the log—sum operation cell circuit. 325 25 .
- the lo-sum operation circuit 325 19 is a 1 og-sum operation circuit 325! Since the same configuration as a detailed description is omitted, 10 g- s um operation cell circuit 32 5 5 Data AG B 104 and rice one enable signal EN 104 supplied from, and, 1 og - s um operation cell circuit 325 6 data AGB 105 and I Ne supplied from the - using enable signal EN 1 05, the operation to be likened to a second round of the tournament Is performed to perform one log-sum operation in the cumulative addition operation of the log-sum operation.
- the 10 g-sum arithmetic circuit 325 19 supplies the calculated data AGL as data AGB 202 to the 10 g-sum arithmetic cell circuit 325 26 which performs an operation analogous to the third round in the winning game, and also enables the enable signal.
- EN202 is supplied to the log-sum operation cell circuit 325 2 ⁇ .
- 10 g-sum arithmetic circuit 325 2 Has a configuration similar to that of the log-sum operation circuit 325, and therefore detailed description is omitted.
- the data A GB106 supplied from the 10 g-sum operation cell circuit 325 and the enable signal EN 106, as well as, 10 g- s um operation cell circuit 325 8 de Isseki AGB 107 and rice supplied from - using enable signal EN 107, by performing an operation to be likened to a second round of the tournament, One log-sum operation in the cumulative addition operation of the log-sum operation is performed.
- 10 g—sum arithmetic circuit 325 2 Supplies the computed data AGL as de Isseki AGB 203, together with the supply to the log- s um operation cell circuit 325 26, rice one enable signal EN 203 to log- s um operation cell circuit 325 26.
- AGB109 and the enable signal EN109 supplied from the United States and performing an operation analogous to the second round in the winning game one of the operations in the cumulative addition operation of the 10 g-sum operation is performed. Perform log-sum operation.
- log- s um operation circuit 325 21 is a data AGB 204 de Isseki AG L calculated, and supplies to the 10 g- sum operation cell circuit 325 27 to perform the operation to be likened to the third round of the tournament, Ineburu
- the signal EN204 is supplied to the 10 g-sum operation cell circuit 325 27 .
- the lo-sum operation circuit 325 22 is a 1 og-sum operation circuit 325! Since the configuration is the same as that described above, the detailed description is omitted, but the data AGB 110 supplied from the 10 g—sum operation cell circuit 325 and the enable signal EN 110 and the 10 g ⁇ s um Operation cell circuit 325 12 By using the enable signal EN111 and performing an operation analogous to the second round in the winning game, one log-sum operation in the cumulative addition operation of the log-sum operation is performed.
- the 10 g—sum operation circuit 325 22 supplies the calculated data AGL as data AGB 205 to the log—sum operation cell circuit 325 27 , and also outputs the enable signal EN 205 to the log—sum operation cell circuit 325. Supply 27 .
- the lo-sum operation circuit 325 23 is a 1 og-sum operation circuit 325! Since the configuration is the same as that described above, a detailed description will be omitted. However, the data AGB 112 supplied from the 10 g—s um operation cell circuit 32 5 13 and the enable signal EN 112, and 10 ⁇ Sum operation cell circuit 325! Using the data AGB113 supplied from 4 and the enable signal EN113, an operation similar to the second round in the winning game can be performed, and one log in the cumulative addition operation of the 10g-sum operation can be obtained. Perform one sum operation.
- the 10 g—sum operation circuit 325 23 supplies the calculated data AGL as data AGB 206 to the 10 g—sum operation cell circuit 32 5 28 that performs an operation analogous to the third round in the winning game. supplying rice one bull signal EN 206 to log- s um operation cell circuit 325 28.
- the lo-sum operation circuit 325 24 Since the lo-sum operation circuit 325 24 has the same configuration as the 1 og-sum operation circuit 325, the detailed description is omitted, but the data supplied from the 10g-sum operation cell circuit 32 5 15 AGB 2nd round in the winning game using the 11.4 and enable signal EN114 and the data AGB115 supplied from the 10-sum operation cell circuit 325 and ⁇ and the enable signal EN115.
- the 10 g—sum arithmetic circuit 325 24 supplies the calculated data AGL as the data AGB 207 to the log—sum arithmetic cell circuit 325 28, and outputs the enable signal EN 207 to the log—sum. and supplies the operation cell circuit 325 28.
- the 10 g—sum operation circuit 325 25 Since the 10 g—sum operation circuit 325 25 has the same configuration as the log—sum operation circuit 325, the detailed description is omitted, but the data supplied from the 10 g—sum operation cell circuit 32 5 17 AGB 200 and enable signal EN 200, and 1 o g-s um operation cell circuit 325 i 8 Using the AGB 201 and enable signal EN 201 supplied from the — Performs one log-one sum operation in the cumulative addition operation of the sum operation.
- the 10 g—sum arithmetic circuit 325 25 supplies the calculated data AGL as data AGB 300 to the 10 g—sum arithmetic cell circuit 325 29 that performs an operation analogous to the fourth round in the winning game. supplying an Bull signal EN300 to log- s um operation cell circuit 325 29.
- the log-sum arithmetic circuit 32 5 26 is a 1 og-sum arithmetic circuit 32 5! Since the same configuration as a detailed description is omitted, 10 g- s um operation cell circuit 32 5 19 data AGB 202 is supplied from and rice one enable signal EN 202, as well as, 1 o g- s um operation cell circuit 32 5 2. Using the data supplied from the AGB 203 and the enable signal EN 203, one operation of the log-sum operation in the cumulative addition operation of the log-sum operation is performed by performing an operation analogous to the third round in the winning game. I do.
- the 10 g—sum arithmetic circuit 325 26 supplies the calculated data AGL as data AGB 301 to the log—sum arithmetic cell circuit 32 5 29 , and also outputs the enable signal EN 301 to the log—sum It is supplied to the arithmetic cell circuit 325 29 .
- the log-sum arithmetic circuit 325 27 has the same configuration as the 1 og-sum arithmetic circuit 325, the detailed description is omitted, but it is supplied from the 1 og-sum arithmetic cell circuit 32 5 21 that data AGB 204 and rice one enable signal EN 204, as well as, 10 g - s um by using an arithmetic cell circuit 32 5 22 data a GB 205 supplied from and I enable signal EN 205, in the third round of the tournament By performing an operation that can be compared, one log-one sum operation in the cumulative addition operation of the 10 g-sum operation is performed.
- the 10 g-sum arithmetic circuit 325 27 performs the operation analogous to the fourth round in the winning game, using the calculated data AGL as the data AGB 302, and a 1 og-sum arithmetic cell circuit 325 3 .
- the enable signal EN302 is supplied to the log-sum operation cell circuit 325 3 . To supply.
- the lo-sum operation circuit 325 28 has the same configuration as the 1 og-sum operation circuit 325, and therefore detailed description is omitted, but the 10 g-sum operation cell circuit 32 5
- 10 g- s um arithmetic circuit 32 5 28 The calculated data AG L as data AGB 303, log- s um operation cell circuit 325 3.
- Be supplied to Rutotomoni supplies rice one enable signal EN 303 to 10 g- s um operation cell circuit 325 30.
- the 1 og-sum operation circuit 325 29 has the same configuration as the 1 og-sum operation circuit 325 i, the detailed description is omitted, but it is supplied from the 1 og-sum operation cell circuit 32 5 25 Using the AGB 301 and enable signal EN301 supplied from the AGB 300 and enable signal EN 300 and the 1-sum operation cell circuit 325 2 ⁇ By performing an operation that can be compared to the fourth round in the above, one 10g-1sum operation in the cumulative addition operation of the 1 og-sum operation is performed.
- the 10 g—s um calculation circuit 325 29 performs the operation analogous to the fifth round in the winning match, here the final match, using the calculated data AGL as the data AGB 400 325 31
- the enable signal EN 400 is supplied to the log-sum operation cell circuit 325 31 .
- 10 g—sum arithmetic circuit 325 3 Is a log-sum operation circuit 325! Since the configuration is the same as that described above, the detailed description is omitted, but the data AGB 302 and enable signal EN 302 supplied from the 10 g-sum operation cell circuit 32 5 27 and the 10 g-sum Using the data AGB 303 and enable signal EN 303 supplied from the arithmetic cell circuit 325 28, by performing an operation analogous to the fourth round in the winning game, one of the operations in the cumulative addition operation of the log-sum operation is performed. log-Performs a sum operation. 10 g—sum arithmetic circuit 325 3 .
- the 1 og-sum operation circuit 325 31 has the same configuration as the 1 og-sum operation circuit 325 i, the detailed description is omitted, but the 1 og-sum operation circuit 325 31 is supplied from the 10 g-sum operation cell circuit 32 5 29 , And the enable signal EN400, and the 10 g-sum operation cell circuit 325 3 .
- the AGB 401 and enable signal EN 401 supplied from the PC an operation analogous to the final match in the winning game is performed, and the log-s in the cumulative addition operation of the log-sum operation is performed.
- the 1 og-sum arithmetic circuit 325 31 does not output the calculated enable signal EN500, but outputs the calculated data AGL as data AGB500.
- the data AGB 500 is supplied to the I input / output circuit 3 13 as data L00.
- Such a log-sum operation circuit 3 1 2 using the data AGB and the enable signal ENS 0, wins a battle based on the enable signal corresponding to each branch on the trellis.
- the lo-sum operation circuit 3 1 2 2 is a 1 og-sum operation circuit 3 12! Since the configuration is the same as that of, the detailed description is omitted, but using the data AGB and the enable signal E NS 1, as with the 10 g-sum arithmetic circuit 3 1 2, By performing an operation that can be compared to a winning match based on the enable signal corresponding to the branch, for example, the cumulative addition operation of a 10 g-sum operation in which the input of the branch on the trellis is "1" is performed, and L01 is calculated.
- the 10 g—sum operation circuit 3 1 2 2 supplies the calculated data L 01 to the I-person calculation circuit 3 13.
- the 10 g—sum arithmetic circuit 3 1 2 3 also has the same configuration as the 1 og—sum arithmetic circuit 3 1 2 i, so a detailed description is omitted.
- the 10 g—sum arithmetic circuit 3 1 2 3 supplies the calculated data L 10 to the I-input calculation circuit 3 13.
- the 1 og—sum operation circuit 3 1 2 has the same configuration as the 1 og—sum operation circuit 3 1 2, detailed description is omitted, but the data AGB and the enable signal EN S 3 and the 10 g-sum operation circuit 3 1 2, as in the trellis, by performing an operation analogous to a winning match based on the enable signal corresponding to each branch on the trellis.
- the input of the upper branch is "1”, and the cumulative addition of the 10g-sum operation is performed to calculate data L11.
- the 10 g—sum operation circuit 3 1 2 4 supplies the calculated data LI 1 to the ⁇ calculation circuit 313.
- the 10 g—sum arithmetic circuit 3 1 2 5 is also a 1 og—sum arithmetic circuit 3 1 2! Since the configuration is the same as that described above, the detailed description is omitted, but using the data AGB and the enable signal ENS20, the same as the 1 og-sum arithmetic circuit 3 1 2, By performing an operation that can be compared to a winning match based on the enable signal corresponding to each branch, for example, the cumulative addition of a 10 g—sum operation in which the input of the branch on the trellis is “0” is performed, and the data L Calculate 20. log- s um calculation circuits 3 1 2 5 supplies the calculated de Isseki L 20 to I's calculation circuit 3 1 3.
- the 1 og—sum arithmetic circuit 3 1 2 6 also has the same configuration as the 1 og—sum arithmetic circuit 3 1 2, so detailed description is omitted, but the data AGB and the enable signal ENS 2 1 Using 10 g—sum arithmetic circuit 3 1 2! Similarly to, by performing an operation that can be compared to a winning match based on the enable signal corresponding to each branch on the trellis, for example, the input of the branch on the trellis is "10, 10 g-sum performs cumulative addition operation of arithmetic to calculate the data L 2 1. 10 g- s um arithmetic circuit 3 1 2 6 supplies the calculated data L 2 1 to I lambda calculation circuit 313.
- ⁇ ⁇ calculation circuit 3 13 has three differentiators 324 324 2 and 324 3 .
- the differentiator 324> receives the data L 0 0 supplied from the 10 g—sum operation circuit 3 12, and the data L 0 1 supplied from the 10 g—sum operation circuit 3 1 2 2 Take the difference of Difference machine 324!
- the data LM0 calculated according to is subjected to, for example, 2's com notation conversion.
- the differentiator 324 2 calculates the difference between the data L 1 0 supplied from the 10 g—sum operation circuit 3 12 3 and the data L 11 supplied from the 10 g—sum operation circuit 3 1 2 4 Take
- the data LM 1 calculated by the differentiator 324 2 is, for example, 2's complement notation conversion. Is applied.
- the differentiator 324 3 is composed of the data L 2 0 supplied from the 1 og—sum operation circuit 3 1 2 4 and the data L 2 1 supplied from the 10 g—sum operation circuit 3 1 2 6 Take the difference of Data LM2 calculated by differentiator 324 3, for example, complement notation conversion or the like of 2 is performed.
- the I ⁇ calculation circuit 3 13 is supplied from each of the 10 g—sum calculation circuits 3 1 2 3 1 2 2 , 3 1 2 3 1 2 4 , and is called a so-called straight binary.
- the notation data L00, L01, L10, and L11 are bundled and output as a logarithmic soft output SLM calculated in symbol units.
- I's calculation circuit 3 1 3, differentiator 324 324 2, 3 24 3 de it it is a 2's complement representation calculated by the Isseki LMO, bundling LM 1, LM2, calculated in bits Output as log soft output BLM.
- the soft-output calculation circuit 16 1 configured as described above performs an operation analogous to a winning game using an enable signal, thereby performing a 10 g-sum operation corresponding to the input of each branch on the trellis.
- the logarithmic soft output I ⁇ can be calculated in symbol units or bit units, and output as log soft outputs SLM and BLM, respectively. These log soft outputs SLM and BLM are supplied to an external information calculation circuit 163, an amplitude adjustment and crib circuit 164, and a hard decision circuit 165.
- the reception value or prior probability information separation circuit 16 2 separates the reception value or prior probability information from the reception data and delay reception data DAD output from the delay storage circuit 155 and subjected to a predetermined delay. It is something to take out.
- the reception value or prior probability information separation circuit 162 is input based on the reception value format information CRTY supplied from the control circuit 60 and the input bit number information IN supplied from the code information generation circuit 151. Separated delayed reception data DAD.
- the reception value or prior probability information separation circuit 162 can be realized as having four selectors 341, 342, 343, 344, for example, as shown in FIG.
- the selector 341 selects one of the delayed reception data DAD 3 and DAD 4 of the delay reception data DAD based on the input bit number information IN. Specifically, when the number of input bits to the element encoder is “1”, the selector 341 selects the delayed reception data DAD4. The selector 341 outputs the selected data as delayed reception data DAS.
- the selector 342 selects one of the delayed received data DAD0 of the delayed received data DAD and the delayed received data DAS supplied from the selector 341, based on the received value format information CRTY. Specifically, the selector 342 selects the delayed reception data DAD 0 when the reception value format information CRY indicates external information. The selector 342 outputs the selected data as delayed reception data P D0.
- the selector 343 selects one of the delayed reception data DAD 1 and DAD 4 of the delay reception data D AD based on the reception value format information C RTY. Specifically, the selector 343 selects the delayed reception data DAD 1 when the reception value format information CRY indicates external information. The selector 343 outputs the selected data as the PD1.
- the selector 344 selects one of the delay reception data DAD2 and DAD5 of the delay reception data DAD based on the reception value format information CRTY. Specifically, the selector 344 selects the delayed reception data DAD2 when the reception value format information C RTY indicates external information. The selector 344 outputs the selected data as delayed reception data PD2.
- Such a reception value or prior probability information separation circuit 162 bundles the delay reception data DAD O, DAD 1, DAD 2, and DAD 3 among the input delay reception data DAD, so-called offset binary. Notation is output as the delayed reception value DRC, and the delayed reception data DAS, DAD4, and DAD5 are bundled and output as the delay prior probability information DAP, and the delayed reception data PD0, PD1, and PD2 are output. Bundled and output as delayed external information DEX.
- the delayed reception value DRC is supplied to the external information calculation circuit 163 and the hard decision circuit 165, and the delay prior probability information DAP is supplied to the external information calculation circuit 163, and the delayed external information DEX is left as it is. It is supplied to the selector 1 2 0 2 as an information sdex.
- the external information calculation circuit 163 is a logarithmic soft output supplied from the soft output calculation circuit 161.
- the external information 0E is calculated using the SLM or the log soft output BLM and the received value or the delayed received value DRC or the delayed prior probability information DAP supplied from the prior probability information separating circuit 162.
- the external information calculation circuit 163 calculates an information bit external information calculation circuit 350 for calculating external information for an information bit, and calculates external information for an information symbol. It can be realized as having an information symbol external information calculating circuit 351, a code external information calculating circuit 352 for calculating external information for a code, and two selectors 353 and 354.
- Information bit extrinsic information calculation circuit 350 has, for example, three external information computation cell circuit 3 5 51 35 5 2, 355 3. These external information computation cell circuit 3 5 5 ,, 3 55 2 35 5 3, it it, in effect, differentiator (not shown) taking the difference between the log soft-force B LM and delayed a priori probability information DAP Consists of
- External information calculation cell circuit 355! Calculates the difference between the log soft output B LM 0 of the log soft output B LM and the delayed prior probability information DAP 0 of the delayed prior probability information DAP, and performs amplitude adjustment and clipping on the difference value. And then perform offset binary notation conversion and the like, and then output as external information EX0.
- the external information calculation cell circuit 35 5 2 calculates the difference between the log soft output B LM 1 of the log soft output B LM and the delayed prior probability information DAP 1 of the delayed prior probability information DAP, and calculates the difference value. After that, amplitude adjustment and clipping are performed, offset binary notation conversion is performed, and then output as external information EX1.
- the external information calculation cell circuit 3 5 5 3 calculates the difference between the log soft output B LM 2 of the log soft output B LM and the delayed prior probability information DAP 2 of the delayed prior probability information DAP, and calculates this difference.
- the value is subjected to amplitude adjustment and clipping, offset binary notation conversion, etc., and then output as external information EX2.
- Such an information bit external information calculation circuit 350 calculates, for example, three systems of external information EX 0, EX 1, and EX 2 in bit units, and bundles these external information EX 0, EX 1, and EX 2. It is supplied to the selector 353 as external information EXB.
- Information Shinporu external information calculating circuit 3 5 1 has, for example, four external information computation cell circuits 35 61, 3 56 2, 3 56 3, and 3 5 6 4, and a normalizing circuit 3 57. These Of each section, the external information computation cell circuit 3 5 61, 3 56 2, 35 6 3 35 6 4, it it, like the external information computation cell circuits 35 5 35 5 2 3 5 5 3, substantially More specifically, it comprises a not-shown differentiator that calculates the difference between the log soft output SLM and the delayed prior probability information DAP.
- the external information calculation cell circuit 356 i calculates a difference between the log soft output S LM 0 of the log soft output S LM and data having a predetermined value M, and performs amplitude adjustment and After clipping, it is supplied to the normalization circuit 357 as external information ED0.
- External information computation cell circuit 3 56 2 the log soft-output S LM 1 of log soft-output S LM, issued calculate the difference between the delay priori probability information DAP 0 of delay priori probability information DAP, the difference value Then, after performing amplitude adjustment and clipping, the information is supplied to the normalization circuit 357 as external information ED1.
- the external information calculation cell circuit 356 3 calculates a difference between the log soft output S LM 2 of the log soft output S LM and the delay prior probability information DAP 1 of the delay prior probability information DAP, and calculates the difference value. On the other hand, after performing amplitude adjustment and clipping, it supplies it to the normalization circuit 357 as external information ED2.
- External information computation cell circuit 3 56 4 issues calculated logarithmic soft-output S LM 3 of log soft-output S LM, the difference between the delay priori probability information DAP 2 of the delay priori probability information DAP, the difference value Is subjected to amplitude adjustment and clipping, and then supplied to a normalization circuit 357 as external information ED3.
- the normalization circuit 357 external information computation cell circuit 3 5 61 35 6 2 3 56 3, 35 6 4 extrinsic information ED 0 calculated by, ED I, of ED 2, ED 3
- ED I ED 1
- ED 2 ED 3
- Clipping is performed according to the dynamic range, and the value of the external information for one symbol is differentiated from the value of the external information for all other symbols. Is performed.
- the normalizing circuit 357 outputs the normalized external information as external information EX 0, EX 1, EX 2.
- Such an information symbol external information calculation circuit 351 calculates, for example, three systems of external information EX O, EX 1, and EX 2 on a symbol basis, and bundles these external information EX 0, EXI, and EX 2 to form the external information. It is supplied to the selector 353 as EXS.
- Code extrinsic information calculation circuit 3 52 has, for example, three external information computation cell circuits 3581, 35 8 2 35 8 3. These external information computation cell circuit 3 58 358 2, 358 3, it it, like the external information computation cell circuit 3 551, 3 55 2, 3 5 5 3, substantially, the log soft-output B LM It consists of a not-shown differentiator that calculates the difference from the delayed reception value DRC.
- External information calculation cell circuit 3 58! Calculates the difference between the logarithmic soft output B LM 0 of the log soft output B LM and the delayed received value AP S 0 of the delayed received value DRC, and adjusts the amplitude and clips the difference. After performing offset binary notation conversion, etc., it is output as external information EX0.
- the external information calculation cell circuit 3 58 2 calculates the difference between the log soft output B LM 1 of the log soft output B LM and the delayed received value AP S 1 of the delayed received value DRC, and calculates the difference value. After performing amplitude adjustment and clipping, and further performing offset binary notation conversion, it outputs it as external information EX1.
- the external information calculation cell circuit 3 58 3 calculates the difference between the log soft output B LM 2 of the log soft output B LM and the delayed received value APS 2 of the delayed received value DR C, and calculates the difference.
- the value is subjected to amplitude adjustment and clipping, subjected to offset binary notation conversion, and the like, and then output as external information EX2.
- Such a code external information calculation circuit 352 calculates, for example, three systems of external information EX 0, EX 1, EX 2, and bundles these external information EX 0, EX 1, EX 2 to obtain the external information EX C To selector 354.
- the selector 353 receives the external information EXB supplied from the information bit external information calculation circuit 350 and the external information EXS supplied from the information symbol external information calculation circuit 351, Select one of them. Specifically, the selector 353 sets the prior probability information format information CAP P If it is a unit, select EXTERNAL INFORMATION EXS. The selector 353 supplies the external information ES obtained by the selection to the selector 354. The selector 354 selects one of the external information ES supplied from the selector 353 and the external information EXC supplied from the code external information calculation circuit 352 based on the output data selection control signal CITM. select. Specifically, selector 3
- Numeral 54 selects the external information EXC when the output data selection control signal CITM indicates that information on the code is to be output.
- the selector 354 outputs external information OE obtained by the selection to the outside.
- the external information calculation circuit 163 calculates the external information OE using the input logarithmic soft output SLM or logarithmic soft output BLM and the delay reception value DCR or the delay prior probability information DAP.
- the external information OE is supplied as it is to the selector 120 as external information SOE.
- the amplitude adjusting and clipping circuit 164 adjusts the amplitude of the logarithmic soft output SLM in symbol units and clips to a predetermined dynamic range, and the logarithmic soft output in bit units. And a circuit for adjusting the amplitude of the BLM and clipping it to a predetermined dynamic range.
- the amplitude adjustment and clipping circuit 164 comprises an output data selection control signal CITM supplied from the outside and a control circuit
- the amplitude of each of the logarithmic soft outputs S LM and B LM is adjusted, and one of the data clipped to a predetermined dynamic range is selected.
- One is output as logarithmic soft output 0 L after amplitude adjustment.
- the logarithmic soft output OL is supplied to the selector 120 as a soft output S0L as it is.
- the hard decision circuit 165 makes a hard decision on the logarithmic soft outputs SLM and BLM, which are the decoded values, and also makes a hard decision on the delayed received value DRC.
- the hard decision circuit 16 5 includes the output data selection control signal CITM supplied from the outside, the received value format information CRTY, the prior probability information format information CAPP supplied from the control circuit 60, and the signal point arrangement.
- logarithmic soft outputs S LM and B LM and delayed reception value DRC are hard-decided.
- the signal point constellation information CSIG shall consist of eight sets of signal point constellation information CSIG 0, CSIG 1, CSIG 2, CSI G3, CSI G4, CS IG 5, CS IG 6, and CS I G7. .
- the hard decision circuit 165 includes an inverter 36 °, a minimum symbol calculation circuit 361 for calculating a symbol having a minimum value, and a selector 369 described later.
- Control signal generation circuit 368 for generating a control signal for controlling the selection operation by the selector, selectors 369 and 371, and the case where the encoding device 1 performs encoding by TTCM or SCT CM. It can be realized as having an I / Q de-mapping circuit 370 that de-mubs I / Q values.
- the receiver 360 inverts a predetermined group of bits of the logarithmic soft output B LM supplied from the soft output calculation circuit 161 and expressed in two's complement notation as decoded bit hard decision information B HD. Output.
- the minimum symbol calculation circuit 361 can be realized as having, for example, three comparison circuits 362, 364, 366 and three selectors 363, 365, 367.
- the comparison circuit 362 compares the magnitude relationship between the log soft outputs S LM0 and S LM1 of the log soft outputs S LM supplied from the soft output calculation circuit 16 1 and expressed in a straight binary notation.
- the comparison circuit 362 supplies the control signal SL0 indicating the obtained magnitude relationship to the selector 367 and also supplies the selector 363 as a control signal for selection.
- the selector 363 selects one of the logarithmic soft outputs SLM0 and SLM1 having a smaller value based on the control signal SLO supplied from the comparison circuit 362.
- the selector 363 supplies the selected data S SLO to the comparison circuit 366.
- the comparison circuit 364 compares the magnitude relation of the log soft outputs S LM2 and S LM3 among the log soft outputs S L supplied from the soft output calculation circuit 16 1.
- the comparison circuit 364 supplies a control signal SL1 indicating the obtained magnitude relationship to the selector 367 and also supplies the selector 365 as a control signal for selection.
- the selector 365 based on the control signal SL1 supplied from the comparison circuit 364, Select the logarithmic soft output S LM2 or S LM3 with the smaller value.
- the selector 365 supplies the data SSL 1 obtained by selection to the comparison circuit 365.
- the comparison circuit 366 compares the magnitude relationship between the data SSL0 supplied from the selector 363 and the data SSL1 supplied from the selector 365.
- the comparison circuit 366 supplies the control signal SEL1 indicating the obtained magnitude relation to the selector 366 as a control signal for selection.
- the selector 367 determines which of the control signal SL 0 supplied from the comparison circuit 362 and the control signal SL 1 supplied from the comparison circuit 364 based on the control signal SEL 1 supplied from the comparison circuit 366. Select one or the other. Specifically, the selector 367 selects the control signal SL1 when the value of the data SSL0 is larger than the value of the data SSL1. The selector 367 outputs the selected data as a control signal SEL0.
- Such a minimum symbol calculation circuit 36 1 calculates the minimum value of the logarithmic soft outputs S LM in symbol units and decodes the decoded symbol hard decision information S obtained by bundling the control signals SEL 0 and SEL 1. It is supplied to the selector 369 as HD.
- the selection control signal generation circuit 368 performs selection by the selector 369 based on the output data overnight selection control signal CITM supplied from outside and the prior probability information format information CAPP supplied from the control circuit 60. Generates control signal AIS for controlling operation.
- the selector 369 based on the control signal AIS supplied from the selection control signal generation circuit 368, decodes the decoded bit hard decision information BHD supplied from the inverter 360 and the decoding supplied from the minimum symbol calculation circuit 361. Either one of Simpol hard decision information S HD is selected. More specifically, the selector 369 outputs the output data selection control signal CITM indicating that the control signal AIS outputs information corresponding to an information symbol or an information bit, and indicates that the control signal AIS is a symbol unit in advance. If it is indicated by the CAP P, the decoded symbol hard decision information S HD is selected. The selector 369 outputs the selected data as decoded value hard decision information DHD1.
- the hard-decision circuit 165 uses these components to generate decoded bit hard-decision information BHD.
- the decoded symbol hard decision information SHD is obtained, and the decoded value hard decision information DHD 1 selected by the selector 369 is output as the decoded value hard decision information DHD.
- This decoded value hard decision information DHD is output to the outside as decoded value hard decision information SDH.
- the hard decision circuit 165 uses the inverter 360 to obtain the decoded bit hard decision information BHD, but this is due to the data notation. That is, as described above, the decoded bit hard decision information BHD is obtained on the premise of the logarithmic soft output BLM expressed in two's complement notation. Therefore, the hard decision circuit 165 uses an inverter 360 to convert an inverted bit obtained by inverting a predetermined group of bits, specifically, the most significant bit, of the log soft output BLM. By using this, it is possible to make a hard decision on the log soft output BLM calculated in bit units.
- the I / Q demapping circuit 370 includes, for example, a look-up table 372 for storing a table for demapping, and seven selectors 373, 374, 375, 376, 377 , 379, 380, and a selection control signal generation circuit 378 that generates a control signal for controlling the selection operation by the selectors 379, 380.
- the lookup table 372 stores a table for demapping received values. More specifically, the lookup table 372 stores a boundary value for the I axis on the I / Q plane as a table, as described later. The look-up table 372 stores a boundary corresponding to a combination of the delayed received value IR corresponding to the in-phase component and the delayed received value QR corresponding to the quadrature component among the delayed received values DRC represented in the offset binary notation. The value is read from the table and supplied to the selection control signal generation circuit 378 as, for example, four-system boundary value data BDR0, BDR1, BDR2, and BDR3.
- the selector 373 selects one of the signal point arrangement information CSIG 2 and CSIG 6 based on the delayed reception value QR. Specifically, if the delayed reception value QR indicates a positive value, the selector 373 selects the signal point arrangement information CSIG2. The selector 373 compares the selected data with the signal point arrangement information SSSS 0. And supplies it to the selector 380.
- the selector 374 selects one of the signal point arrangement information CSIG3 and CSIG5 based on the delayed reception value QR. Specifically, if the delayed received value QR indicates a positive value, the selector 374 selects the signal point arrangement information CSIG3. The selector 374 supplies the selected data to the selector 376 as signal point arrangement information S S0.
- the selector 375 selects one of the signal point arrangement information CSIG1 and CSIG7 based on the delayed reception value QR. Specifically, when the delayed reception value QR indicates a positive value, the selector 375 selects the signal point arrangement information CSIG1. The selector 375 supplies the selected data to the selector 376 as signal point arrangement information S S1.
- the selector 376 selects one of the signal point arrangement information SSO supplied from the selector 374 and the signal point arrangement information SS1 supplied from the selector 375 based on the delayed reception value IR. Specifically, when the delay reception value IR indicates a positive value, the selector 376 selects the signal point arrangement information SS1. The selector 376 maps the selected data to the signal point arrangement. Information SSS 0 is supplied to selector 379.
- the selector 377 selects one of the data having the predetermined value M and the signal point arrangement information CSIG4 based on the delayed reception value IR. Specifically, the selector 377 selects data having a predetermined value M when the delayed reception value IR indicates a positive value. The selector 377 supplies the selected data to the selector 379 as signal point arrangement information S S S 1.
- the selection control signal generation circuit 378 performs the selection operation by the selector 379 based on the delayed reception value QR and the boundary value data BDR 0, BDR 1, BDR 2, and BDR 3 supplied from the lookup table 372. It generates a control signal SEL 5 for controlling and a control signal SEL 6 for controlling the selection operation by the selector 380.
- the selector 379 selects one of the signal point arrangement information SSS0 and SSS1 based on the control signal SEL5 supplied from the selection control signal generation circuit 378. select.
- the selector 379 supplies the selected data to the selector 380 as signal point arrangement information SSSS1.
- the selector 380 selects one of the signal point arrangement information SSSSS0 and SSSSS1, based on the control signal SEL6 supplied from the selection control signal generation circuit 378.
- the selector 380 supplies the selected data to the selector 371 as the received value hard decision information I RH.
- Such an I / Q demultiplexing circuit 370 obtains the received value hard decision information I RH when the encoding device 1 performs encoding by TTCM or SCTCM.
- the selector 371 is composed of a predetermined bit group of the delayed received value DCR, and the received value hard decision information BRH indicating the hard decision result in offset binary notation, and I / Q Demap circuit 370 selects one of received value hard decision information I RH supplied from demapping circuit 370. Specifically, if the received value format information CR TY indicates that the encoding device 1 performs encoding by TT CM or SCT CM, the selector 37 1 Select Information I RH. The selector 371 outputs the selected data as the received value hard decision information RHD. This received value hard decision information RHD is output to the outside as received value hard decision information SRH.
- the hard decision circuit 165 does not perform the bit inversion processing to obtain the received value hard decision information B RH as in the case of obtaining the decoded bit hard decision information BHD described above. This is due to the notation. That is, as described above, the received value hard decision information BRH is obtained on the premise of the delayed received value D RC that is expressed in offset binary notation. Therefore, the hard decision circuit 1665 makes a hard decision on the delayed received value DRC by making a decision using a predetermined bit group of the delayed received value DRC, specifically, the most significant bit. Can be.
- Such a hard decision circuit 165 makes a hard decision on the logarithmic soft outputs S LM and B LM that are the decoded values to obtain decoded value hard decision information SDH, and makes a hard decision on the delayed received value DRC to make the received value hard decision.
- Information Ask for S RH The decoded value hard decision information S DH and the received value hard decision information SRH are, respectively, the decoded value hard decision information DHD and the received value hard decision.
- the soft output decoding circuit 90 described above receives the soft input decoded received value TSR.
- the Ia calculation circuit 156 and the Ia distribution circuit 157 each time the received value is received, the log likelihood I
- the I calculation circuit 159 calculates the log likelihood for each state at all times. Calculate the likelihood I?
- the element decoder 50 calculates the log soft output I at each time using the log likelihood I, I and Ia calculated by the soft output calculation circuit 161, and calculates the log soft output It outputs I ⁇ to the outside or supplies it to the external information calculation circuit 163.
- the element decoder 50 calculates external information at each time by the external information calculation circuit 163.
- the element decoder 50 can perform soft output decoding to which the Log-BCJR algorithm is applied, using the decoded reception value TSR and the external information or the in-date data.
- the soft-output decoding circuit 90 can perform soft-output decoding on an arbitrary code regardless of the code configuration of the element encoder in the PCC, SCC, TTCM, or SCTCM.
- interleaver 100 will be described in detail. Prior to a description of a specific configuration, a basic design concept of the interleaver 100 will be described.
- the interleaver 100 can perform an interleave process and a dinter-leave process, and can also delay an input received value. Therefore, it is assumed that the in-leaver 100 has a RAM for delaying the input received value and a RAM for performing the in-leave on the input data. As described later, these RAMs are actually shared and are used by being switched according to the mode indicating the code configuration including the type of interleave to be performed. It is.
- a delay RAM is provided by a control circuit (described later) of the interleaver 100 from one RA of a dual port consisting of banks A and B. It is configured to look like M.
- the control circuit cannot simultaneously access the even-numbered address or the odd-numbered address by using the write address used for writing data to the RAM and the read address used for reading out the data all at once.
- the interleaver 100 when an even-length delay is performed using the delay RAM, for example, 0, 1, 2, 3, 4, ',', DL — 2, DL-1, 0, Based on the write address such as 1, 2,... ', The data is stored in each address in the RAM.
- the interoperator 100 for example, based on a read address such as 1, 2, 3, 4, 5,..., DL—1, 0, 1, 2, 3, 3: Data is read from each address.
- the in-leaver-reaver 100 is realized by holding an output delayed by an even-length in a register or the like.
- the RAM for delay is composed of a plurality of RAMs for upper addresses and lower addresses of punctures A and B as shown in FIG. 51, for example.
- the in-all-in-one receiver 100 as shown in FIG. 52, for example, it is necessary to appropriately convert the address generated by the control circuit and give it to each RAM.
- the reason for inverting the most significant bit of the address is to simplify designation of the address when inputting / outputting a plurality of symbols, as described later.
- the RAM for in-leaving is configured so that the control circuit looks like two RAMs consisting of banks A and B as shown in FIG. 53, for example.
- the in-leave-reaver 100 can switch between the interleave process and the din-leave-leave process. Therefore, in the case of the interleaver 1 ⁇ 0, when the interleave processing is performed, the count-up, for example, 0, 1, 2, 3,. , 1, 0, etc., the data is stored in each address in the RAM as the write bank A based on the sequential write address generated by counting down. Then, in the interleaver 100, the data is read from each address in the RAM as the read bank B based on the random read address.
- the address storage circuit 110 basically has a read address data which is, for example, a three-system random address data based on the sequential address data IAA supplied from the data receiver 100. Overnight AD A0, ADA1, ADA2 shall be output. In this way, by providing a plurality of systems of read address data ADA from the address storage circuit 110 to the interleaver 100, the interleaver 100 can store up to three symbol data at maximum. Multiple types of interleaving can be performed.
- the interleaver 100 uses three systems from the address storage circuit 110 to perform random interleave on one symbol input data.
- the interleaving is performed using the read address data AD A0 of the read address data ADA0, ADA1, and ADA2.
- a random in-the-night leave will be referred to as a random in-the-night leave.
- the interleaver 100 stores the address.
- Read address data ADA 0, ADA of three systems of read address data A DA 0, AD A 1, and ADA 2 from circuit 110 Use 1 to perform interleaving.
- such an in-leave is referred to as an inline in-leave.
- the interleaver 100 retains the combination of each bit with respect to the input data of two symbols, that is, the interleaver 100 has the same address for each symbol.
- the read address data AD A0 and ADA 1 of the three systems of read address data ADA 0 and ADA 1: ADA 2 from the address storage circuit 110 are used.
- the interleaver 100 when random interleaving is applied to the input data of three symbols, the interleaver 100 reads the three sets of read address data from the address storage circuit 110. Evening Perform interleaving using all of ADA0, ADA1, and ADA2.
- Fig. F when performing in-line interleaving on the input data of three symbols, as shown in Fig. F, three read addresses from the address storage circuit 110 are provided. Interleaving is performed using all of the data ADA 0, AD A 1, and ADA 2.
- the interleaver 100 when performing pairwise in-leave with respect to the input of three symbols, performs three operations from the address storage circuit 110. Performs interleave using all of AD A0, AD A1 and ADA2.
- the interleaver 100 can perform a plurality of types of read / write operations using a plurality of systems of read address data ADA provided from the address storage circuit 110.
- the plurality of types of interleaving includes, of course, a plurality of types of interleaving that perform the reverse of the interleaving.
- the interleaver 100 has a plurality of RAMs, and realizes a plurality of types of interleave by appropriately selecting and switching a RAM to be used according to the type of the interleave.
- the interleaver 100 capable of performing such an interleave process or an interleave process is configured as shown in FIG. 56, for example.
- the interleaver 100 includes a control circuit 400 for performing various processes such as address generation, a delay address generation circuit 401 for generating a delay address, and an odd-length delay compensation circuit 402 for compensating for an odd-length delay.
- Control circuit 400 stores circuit 4071, 407 2, ⁇ ⁇ ⁇ , 407 16 against a controls the writing and / or reading data, Inta one rib ⁇ supplied from selector evening 1 20 5
- a start position signal TIS is input
- a write address and a read address used for interleaving or interleaving are generated.
- the control circuit 400 delays by the interleave mode signal CD IN supplied from the outside, and the in-leave-leave-length information CI NL and the in-leave-leave length supplied from the control circuit 60.
- a write address and a read address are generated based on the operation mode information CB F indicating the operation mode information.
- the control circuit 400 supplies the interleaved address conversion circuit 403 with the generated write address data IWA, which is the generated sequential address data. Further, the control circuit 400 supplies the generated sequential address data IAA to the address storage circuit 110 and also supplies it to the interleave address conversion circuit 403 as an interleave long delay read address data IRA.
- the control circuit 400 includes the termination position information CNFT, the termination period information CNFL, the termination state information CNF D, the puncture cycle information CNE, and the puncture pattern information supplied from the control circuit 60.
- the interleaver no-output position information It generates a CNO and a delay interleave start position signal CDS, and generates an end time information CGT, an end state information CGS, and an erase position information CGE.
- the control circuit 400 outputs the generated information to each other, the interleaver non-output position information I NO, the delay interleave start position signal IDS, and the end time information IG.
- the termination state information IGS and the erasure position information IGE the selector 1 20! .
- the control circuit 400 also supplies the generated output signal C NO to the address selection circuit 405.
- the write address data I WA which is a sequential address data generated by the control circuit 400, is composed of the in-leave mode signal CDIN and the in-leave signal 100. evening when there was one that instructs the performing Ribu processing, storage circuit 407 !, 407 2, ⁇ ⁇ ⁇ , but the address data to be used to write de Isseki for 407 16, in the evening one Ribumo one de signal CD iN is, when the Intariba 1 00 was intended to instructs the performing Dintari some processing, storage circuit 4071, 407 2, ⁇ ⁇ ⁇ , to de Isseki read from 407 " Similarly, the sequential address data IAA generated by the control circuit 400 is an interleave mode signal CD IN, and the interleaver 100 is an interleave process.
- inter one leave mode signal CD iN is, if the fin evening one interleaver 1 00 was intended to instructs to perform deinterleaving processing, storage circuit 4071, 40 7 2, - ..., it becomes to read address storage circuit 1 1 0 to the random address data used for data evening writing for 407 16.
- the control circuit 400 When generating a write address and a read address, the control circuit 400 generates sequential address data by counting up using a counter (not shown). The read address counter is provided separately, as described later.
- the delay address generation circuit 401 generates delay address data based on the instantaneous leave length information CINL supplied from the control circuit 60.
- the delay address generation circuit 401 converts the generated write address data DWA, which is the generated write address data, and the read address data DRA, which is the read address data, into the delay address conversion circuit. Supply to 404.
- the odd length delay compensation circuit 402 is provided for compensating for the odd length delay.
- the interleaver 100 is configured using two banks of RAMs when performing the delay.
- the delay length that is, the interleave length is changed. Data delay can be achieved by using two banks of RAM with half the number of words for the time slot.
- the delay length is limited to an even number in this case. Therefore, the odd-length delay compensation circuit 402 is provided to cope with the odd-length delay, and performs an even-length delay based on the interleave length information CINL supplied from the control circuit 60.
- the delay of the data TDI by one minute of the delay by the RAM and one time slot by the register are performed. Select the TDI, which is the target of the delay, so that the delay is performed.
- the odd-length delay compensation circuit 402 is assumed to have a data TDI composed of six data lines TDIO, TDI1, TDI2, TDI3, TDI4, and TDI5, for example, as shown in FIG.
- TDI data lines
- TDI1, TDI2, TDI3, TDI4, and TDI5 data lines
- FIG. 57 there are 6 registers 4 1 0 4 1 02, 4 1 0 3 , 4 1 04, 4 1 0 5 , 4 1 0 6 and 6 selectors 4 1 1 4 1 1 2 , 4 11 3 , 4 11 4, 4 11 5 and 4 11 6 can be realized.
- the register 410 When the data TDI0 is input, the register 410 holds this data TDI0 for one instant slot. The register 410i supplies the held data DDD0 to the selector 4111i.
- Register 4 1 0 3 inputs the data TDI 2, it holds this data TD I 2 only 1 evening Imusuro' preparative min. Register 4 1 0 3 supplies the held data DDD 2 in cell Lek evening 4 1 1 3.
- register 410 When the register 410 inputs the data TDI 3, it holds the data TDI 3 for one im slot. Register 4 1 0 supplies the de Isseki DDD 3 held in the selector 4 1 1 4.
- Register 4 1 0 5 the data input TDI 4, holds the de Isseki TD I 4 only 1 evening Imusuro' preparative min. Regis evening 4 1 0 5 supplies the held data DD D 4 to selector 4 1 1 5.
- Register 4 1 0 6 inputs the data TD I 5, holds this data TD I 5 only 1 data Imusuro' preparative min. Register 4 1 0 beta supplies data DDD 5 held in selector 4 1 1.
- Selector 4 1 1! Selects one of the data DDD0 supplied from the register 410i and the data TDI0 based on the interleave length information CINL. Specifically, the selector 4 1 1! Selects the data TDI0 if the interleave length is an even number. The selector 411i supplies the selected data DS0 to the input data selection circuit 406 as data D0. This selector 4 1 1! It goes without saying that the interleave length information CINL input to the sub-unit is actually the least significant bit of the bit string representing the interleave length information CINL.
- the selector 4 1 1 based on the fin evening one sleeve length information CI NL, the data DDD 1 supplied from the register evening 4 1 0 2, of a data TD I 1, selects either. Specifically, if the interleave length is an even number, the selector 4 1 1 2 selects the data TDI 1. The selector 4 1 1 2 supplies the selected data DS 1 to the input data selection circuit 406 as data D 1. Incidentally, Inta one sleeve length information CI NL inputted to the selector 4 1 1 2 may actually be sufficient at the lowest bit Bok bits string representing the in-evening one sleeve length information CI NL is Iuma not.
- the selector 4 1 1 on the basis of the Inta one sleeve length information CI NL, and data DDD 2 supplied from the register 4 1 0 3, of the Isseki de TD I 2, selects either. Specifically, the selector 4 1 1 3, when Inta one rib length is an even length, selects the data TDI 2. The selector 4 1 1 3 supplies data DS 2 selected, the input data selecting circuit 40 6 as the data D 2. Incidentally, Inta one sleeve length information CI NL inputted to the selector 4 1 1 3, in fact, it is sufficient at the lowest bit of the bit string representing the fin evening one sleeve length information CI NL is had Not a horse.
- the selector 4 1 1 4 on the basis of the fin evening one sleeve length information CI NL selected, the de Isseki DDD 3 supplied from register evening 4 1 0 4, of the data TD I 3, either I do. Specifically, the selector 4 1 1 4 selects the data TDI 3 when the length of the data is even. The selector 4 1 1 4 supplies data DS 3 selected, as the data D 3 to the input data selection circuit 406. Incidentally, Inta one sleeve length information CI NL inputted to the selector click evening 4 1 1 4, in fact, it is sufficient with the least significant bi Uz preparative bicycloalkyl Uz preparative column representing the in-evening one sleeve length information CINL Needless to say.
- the selector 4 1 1 based on Inta one sleeve length information CI NL, the data DDD 4 supplied from the register evening 4 1 0 5, of the data TD I 4, selects either. Specifically, the selector 4 1 1 5, when inter one leave length is an even length, selects the data TD I 4. The selector 4 1 1 5 supplies data DS 4 selected, the input data selecting circuit 40 6 as data D 4. Incidentally, fin evening one sleeve length information CINL inputted to the selector click evening 4 1 1 5, in fact, suffice two least significant bits in the bit string representing the fin evening one sleeve length information CI NL Needless to say.
- the selector 4 1 1 beta based on the fin evening one sleeve length information CI NL, the data DDD 5 supplied from the register 4 1 0 6, of a data TD I 5, selects either. Specifically, the selector 4 1 1 6, when fin evening one sleeve length is an even length, selects the data TD I 5. The selector 4 1 1 6 supplies the de Isseki DS 5 selected, the input data selecting circuit 40 6 as the de Isseki D 5. In addition, this cell Intaribu length information CINL input Kuta 4 1 1 6, in practice, it is sufficient at the lowest bit Bok bits string representing the in-evening one sleeve length information CI NL is neither Iuma.
- the odd-length delay compensation circuit 402 When the odd-length delay compensation circuit 402 receives the data TDI, the odd-length delay compensation circuit 402 outputs the data TDI without passing through the register in the case of the even-length delay, and outputs the data TDI in the case of the odd-length delay. Evening TDI is held for one time slot by a register and then output.
- the interleave-leave conversion circuit 403 indicates that the interleave mode signal CDIN supplied from the outside and the interleaver type information CINT and the interleave length supplied from the control circuit 60 should be delayed by the interleave length.
- the write address data I WA which is a sequential address data supplied from the control circuit 400
- the address data IRA which is a read signal with a long lead time
- the desired address data is selected from the read address data ADA, which is a random address data supplied from the controller, and converted into address data for in-leave.
- the in-leaf address conversion circuit 403 supplies the address data AA 0, BA O, A A 1, B A 1, A A 2, and B A 2 obtained by the conversion, for example, to the address selection circuit 405.
- the in-leaving address conversion circuit 403 includes, for example, four control signals IOBS, IOBP O, IOBP 1 for instructing the selection operation in the output data selecting circuit 408 based on the input information. , IOBP 2 and supplies these control signals to the output data selection circuit 408.
- the delay address conversion circuit 404 selects desired address data from the delay write address data DWA and the delay read address data DRA supplied from the delay address generation circuit 401, and outputs the delay address data. Convert to overnight.
- the delay address conversion circuit 404 supplies, for example, two systems of address data DAA and DBA obtained by the conversion to the address selection circuit 405. Further, the delay address conversion circuit 404 generates, for example, two control signals DOBS and DOBP for instructing the selection operation in the output data selection circuit 408 based on the input information. The control signal is supplied to the output data selection circuit 408.
- the address selection circuit 405 converts the interleaved address based on the interleaver type information CINT supplied from the control circuit 60 and the interleaver non-output position information CNO supplied from the control circuit 400.
- the address selection circuit 405 selects the selected address data AR 00, AR 01,
- the address selection circuit 405 further includes an interleave address conversion circuit 405 (not shown) that is generated by the control circuit 400 and is not shown.
- the address selection circuit 405 determines a storage circuit 407 407 2 ,
- a write enable signal XWE for 407 16, a memory circuit 4071, 407 2, ⁇ ⁇ ⁇ , clock inhibit for blocking the clock signal for 407 16 (clo ck inhibit) signal the IH, storage circuit 407 i, 407 2 , ... , 407 Generates a partial write control signal PW for writing data to ie as a so-called partial write.
- Adoresu selection circuit 405 these write enable signal XWE, a clock inhibit signal IH and partial la I DOO control signal PW, storage circuit 4071, 407 2, ⁇ ⁇ ⁇ , and supplies to 407 16.
- the input data selecting circuit 40 6 for example, supplied from the selector 1 2 0 4 3 systems of de Isseki TII 0, T ill, with TII 2 is input as data 1 0, II, 1 2, odd number length Data D 0, D 1, D 2, D 3, D 4, and D 5 supplied from the delay compensation circuit 402 are input.
- the input data overnight selection circuit 406 is provided externally. Based on the supplied interleave mode signal CD IN and the interleaver type information c INT and interleaver input / output replacement information CI PT supplied from the control circuit 60, the data 10, II, and 12 are obtained.
- the input data selection circuit 406 inputs the data 10, II, and 12 and these data 10, II, 1 of the two, the memory circuit 40 71, 407 2, ⁇ ⁇ ⁇ , selects the data to be distributed to 407 16.
- the input data selection circuit 406 When delaying the input data, the input data selection circuit 406 inputs the delay data D 0, D 1, D 2, D 3, D 4, D 5, data DO, D 1, D 2, D 3, D 4, of D 5, the memory circuit 407 !, 407 2, ⁇ ⁇ ⁇ , 407, selects the data to be distributed to beta. Input data - evening selection circuit 406, Isseki de selected IR 00, IR 0 1, ⁇ ⁇ ⁇ , the IR 1 5, it it, the memory circuit 407 407 2j ⁇ ⁇ ⁇ , and supplies to 407 16.
- the input data selection circuit 406 has a function of replacing each symbol with each other when performing in-leave on a plurality of symbols. That is, the input data selection circuit 406 has a function of exchanging the order of each symbol for the input data I 0, I I, and I 2 based on the input / output replacement information C I ⁇ .
- the interleaver 100 has these storage circuits 407 407 2 ,. By controlling the writing and / or reading of data with respect to 6, the interleave processing and the interleave processing, and the reception value delay processing can be realized.
- the data IR 00, IR 01,..., IR 15 are collectively referred to as data IR
- the data OR 00, OR 01,. , OR 15 It is collectively called OR.
- Inverter 420 inputs the most significant bit of address data AR and inverts the most significant bit.
- the inverter 420 supplies the inverted bit I AR obtained by the inversion to the selector 421.
- the selector 421 based on the partial write control signal PW supplied from the address selection circuit 405, converts the inverted bit IAR supplied from the inverter 420 and the bit whose value is "0". One of them is selected and output as 1-bit data HPW. Specifically, the selector 421 selects the inverted bit IAR when the partial write control signal PW instructs data writing by the partial write function.
- the data HPW selected by the selector 421 is parallel-converted into, for example, 8 bits and supplied to the RAM 424 as data VIH.
- the selector 422 selects either the most significant bit of the address data AR or the bit whose value is "0" based on the partial write control signal PW supplied from the address selection circuit 405. Then, it is output as 1-bit data LPW. Specifically, the selector 422 selects the most significant bit of the address data AR if the partial write control signal PW instructs data writing by the partial write function.
- the data LPW selected by the selector 422 is parallel-converted into, for example, 8 bits and supplied to the RAM 424 as data VI.
- the data IR is divided into an upper bit and a lower bit and input to the selector 423.
- the selector 423 has a data IR [15: 8] of the upper 8 bits and a data IR [7: 0] is input.
- the selector 423 selects one of the upper bit and the lower bit of the data IR based on the partial write control signal PW supplied from the address selection circuit 405. More specifically, when the partial write control signal PW instructs data writing by the partial write function, the selector 423 selects the lower bits of the data IR.
- the RAM 424 simply writes the data IR and reads the data OR based on the address data AR, but has a partial write function as described above. Therefore, the configuration does not simply input the address data AR and data IR and output the data OR.
- the RAM 424 is supplied with the write enable signal XWE and the clock blocking signal I H supplied from the address selection circuit 405.
- the RAM 424 enters a state in which data can be written.
- Data OH and OL are read from RAM424 based on the address data IA and the data VIH and VIL. These data OH and OL are both supplied to selectors 425 and 426.
- the clock blocking signal I H is input, the RAM 424 stops all operations including writing and / or reading.
- the selector 425 selects one of the data 0H and ⁇ L supplied from the RAM 424 based on the data LPD obtained by delaying the data LPW supplied from the selector 422 with a predetermined delay. Select one and output as SOH. Specifically, the selector 425 selects the data OH when the data LPD is “0”, and selects the data 0 L when the data LPD is “1”. . That is, the selector 425 is provided to determine whether to output higher-order bit data or lower-order bit data in the address direction in consideration of data writing and reading by the partial write function. Things.
- the selector 426 selects one of the data 0H and OL supplied from the RAM 424, based on the data LPD obtained by delaying the data LPW supplied from the selector 422 with a predetermined delay. And outputs it as S 0 L overnight.
- the selector 426 selects the data 0L when the data LPD is "0", and selects the data OH when the data LPD is "1". That is, similarly to the selector 425, the selector 426 considers the writing and reading of data by the partial write function and considers either the data of the upper bit or the data of the lower bit in the address direction. It is provided to determine whether to output.
- the storage circuits 407 407 2 ,..., 407 are based on the address addresses AR 00, AR 01,--. ,..., IR15 writing and data reading OR00, 0R01, ⁇ , 0R15.
- the output data selection circuit 408 includes an interleave mode signal CD IN supplied from outside, the interleaver type information CINT and the interleave input / output replacement information CIPT supplied from the control circuit 60, and an interleave.
- the storage circuit 4071 is based on the control signals IOBS, IOBP0, IOBP1, IOBP2 supplied from the address conversion circuit 403 and the control signals DOBS, DOBP supplied from the delay address conversion circuit 404. , 407 2 ,..., 407 1 ⁇ Select the data to be output from the data OR 00, 0 R 0 1,.
- the output data selection circuit 408 converts the selected data into, for example, three systems of interleaver output data 1100, 1101, and as 1 102, it then supplies to the selector 1 20 7. Further, when the input data is delayed, the output data selection circuit 408 converts the selected data into, for example, six interleave length delay reception values ID ⁇ 0, ID01, ID02, ID03. , I D04, I D05 Is it supplied to the selector 1 2 0 6.
- the output data selection circuit 408 has a function of replacing each symbol with each other when performing dive-leaving on a plurality of symbols, as described later. In other words, the output data selection circuit 408 changes the order of each symbol for the output data output 1 100, 1101, II 02 based on the interleaver input / output replacement information CIPT. Has functions.
- the interleaver 100 described above uses the write address data IWA, which is sequential address data generated by the control circuit 400, and the address select circuit 405 Appropriate storage circuits 407 407 2 ,..., 407 1 ⁇ While distributing the addresses to 1 ⁇ , the input data selection circuit 406 stores the data 10, II, 12 in appropriate storage circuits 40 71, 407 2 ,. ..., it distributes addresses to 407 16, the storage circuits 407 1, 407 2,.,., writes Isseki de 407 16. On the other hand, the interleaver 100 uses read address data ADA, which is random address data read from the address storage circuit 110 based on the sequential address data IA generated by the control circuit 400.
- ADA random address data read from the address storage circuit 110 based on the sequential address data IA generated by the control circuit 400.
- Te Adoresu selection circuit 40 5 Niyotsu, appropriate storage circuit 407 ,, 407 2, ⁇ ⁇ ⁇ , distributes addresses to 407 16, the memory circuit 407 407 2, ⁇ ⁇ ⁇ , stored in 407 16 Read the data that is stored. Then, the interleaver 100 selects the data output from the appropriate storage circuit 407 15 407 2 ,..., 407 16 by the output data selection circuit 408, and outputs the interleaver output data II 00 , 1 101, and I 102. By doing so, the interleaver 100 can perform interleave processing.
- the interleave receiver 100 reads the random address read out from the address memory circuit 110 based on the sequential address data IA generated by the control circuit 400.
- the address selection circuit 405 distributes the address to the appropriate storage circuit 407 l 5 4072, ⁇ , 407 16 and selects the input data overnight.
- the circuit 406 stores the data I 0, II, and 12 in an appropriate storage circuit.
- the interleaver 100 uses the write address data IWA, which is a sequential address data generated by the control circuit 400, and the appropriate storage circuit 4071, 4072 ⁇ by the address selection circuit 405. - distributes the Adoresu to 407 16, the memory circuit 407 407 2> ⁇ ⁇ ⁇ , reads the Isseki de stored in 407 16. Then, the interleaver 1 00, by the output data selection circuit 408, the appropriate storage circuit 407 !, 407 2, ⁇ ⁇ ⁇ , select the data output from 4 07 16, inter one interleaver output data Isseki IIO 0 , 1101, I102. By doing so, the interleaver 100 can perform the din-even-leaving process.
- IWA write address data generated by the control circuit 400
- the interleaver 100 uses the write address data IWA generated by the control circuit 400, and the appropriate memory circuits 4071, 407 by the address selection circuit 405. 2, ⁇ ⁇ ⁇ , 407 as well as distributing the address iota beta, the input data selecting circuit 406, de Isseki D 0, D 1, D 2 , D 3, D 4, D 5 appropriate storage circuits 407! , 407 2 ,..., 407 16 , and these storage circuits 407! , 4072, ⁇ ⁇ ⁇ ⁇ , 4 07 Write the data overnight to 1 ⁇ .
- the input receiver 100 uses the input address data IR ⁇ ⁇ which is sequential address data generated by the control circuit 400 and is read by the address selection circuit 405 by using the address data IR ⁇ ⁇ .
- Do storage circuit 4071, 407 2, ⁇ ⁇ ⁇ distributes Adoresu to 407 16, the memory circuit 4 071, 407 2, ⁇ ⁇ ⁇ , reads the Isseki de stored in 407, 6. Then, inter reaper 1 00, the output data Isseki selection circuit 408, the appropriate storage circuit 4071, 407 2, ⁇ ⁇ ⁇ , select the data output from the 407 16, inter one leave long delayed received value ID 00 , ID 01, ID02, ID03, ID04, ID05. By doing so, the interleaver 100 can delay the input data.
- Element decoder 50 as RAM for data storage circuit 407!
- inter one interleaver 1 00, 407 2, ⁇ ⁇ ⁇ comprising a six RAM with it that of 407 16, as the RAM address And a plurality of RAMs included in the address storage circuit 110.
- the 16 RAMs of each of the storage circuits 407 407 2 ,..., 407 have a storage capacity of 16 bits ⁇ 4096 words
- the address storage circuit 110 is , 14 bits ⁇ 4096 words with six RAMs.
- the RAMs in the storage circuits 407 4 072,..., 407 1 ⁇ are referred to as RAM D 01, D 02,. , RAMA.
- the encoding device 1 performs PCCC with an encoding rate of “1/6 or more” and that the input data capacity is “16 kilowords or less”. .
- the interleaver 100 has, for example, as shown in FIG. 59A, 12 RAMs D 01, D 16 out of 16 RAMs D 01, D 02,..., D 16. 0 2, D 03, D 04, D 05, D 06, D 07, D 08, D 09, D 10, D ll, D 12 are used for the delay, and the remaining The four R AMDs 13, D 14, D 15, and D 16 are used for interleaving.
- the RAM for the address as shown in Fig. C, any four of the six RAMAs may be used. Therefore, as shown in FIG. D, the in-memory switcher 100 and the address storage circuit 110 do not use two RAMAs.
- the interleaver 100 includes R AMD 01, D 02, D 05, D 06, D 09, D 10, D 13, D14 is used as the above-described bank A (A., A,), and RAMD03, D04, D07, D08, Dll, D12, D15, and D16 are used as the above-described bank B. (Used as B, B.
- the interleaver 100 is RAMD 01, D 02, D 05, When data is written to D 06, D 09, D 10, D 13, D 14, RAMD 03, D 04, D 07, D 08, D ll, D 12, D If data is read from 15 and D16 and data is written to RAMD03, D04, D07, D08, Dll, D12, D15, and D16 The data is read from RAMD01, D02, D05, D06, D09, D10, D13, D14.
- R AMD01 and DO2 are respectively delayed as data IROO and IR01 from the input data selection circuit 406 based on the address data AR00 and AR01 supplied from the address selection circuit 405.
- D0 and D1 are supplied and written.
- data of 0 to 4 kilowords out of data D0 and D1 is written to RAMD01, and data of 4 to 8 kilowords is written to RAMD02.
- the RAMD 05 and DO 6 have data IR 04 and IR from the input data overnight selection circuit 406 based on the address data AR 04 and AR 05 supplied from the address selection circuit 405, respectively.
- delay data D 2 and D 3 are supplied and written.
- the stored data is the data OR 0 2, OR 03, OR 06, OR 07, OR 10, OR 11 are read and supplied to the output data selection circuit 408. Note that reading of data is performed based on address data supplied from the address selection circuit 405, as in the case of writing data.
- RAMD 03 and DO 4 Based on the supplied address data AR 02 and AR 03, delay data D 0 and D 1 are supplied and written as data IR 02 and IR 03 from the input data selection circuit 406. At this time, data of 0 to 4 kilowords of the data DO and D1 is written to RAMD03, and data of 4 to 8 kilowords is written to RAMD04. Also, based on the address data AR06 and AR07 supplied from the address selection circuit 405, the RAMD07 and DO8 receive data IR06 and IR07 from the input data selection circuit 406, respectively, for delay. Data D 2 and D 3 are supplied and written.
- the stored data are stored in the data storage OR 00, OR 0 1, It is read as 0 R 04, 0 R 05, OR 08, 0 R 09, and supplied to the output data selection circuit 408.
- the data reading is performed based on the address data supplied from the address selection circuit 405 in the same manner as in the data writing.
- R AMD 13, D 14, D 15, and D 16 each function as a partial-write RAM based on the partial-write control signal PW, and pseudo 8 bits X 8 Acts as a RAM with a storage capacity of 1992 cards.
- R AMD 13 and D 14 are provided with the input data selection circuit 406 from the input data selection circuit 406 based on the address data AR 12 and AR 13 supplied from the address selection circuit 405, respectively.
- Interleave data 10 is supplied and written as 12 and IR 13. At this time, data of 0 to 8 kilowords of data 10 is written to RAMD 13 and RAMD 14 is written of 8 to 16 kilowords. Data is written.
- the stored data is read out from R AMD 15 and D 16 as data OR 14 and 0 R 15, respectively, and supplied to output data selection circuit 408. Note that the data reading is performed based on the address data supplied from the address selection circuit 405, as in the data writing.
- R AMD 15 and D 16 each have data IR 15, from input data selection circuit 406 based on the address data AR 14 and AR 15 supplied from address selection circuit 405.
- the data I 0 for interleaving is supplied and written as IR 16.
- the data of 0 to 8 kilowords out of the data 10 is written to the RAMD 15 and the data of 8 to 16 kilowords is written to the RAMD 16.
- the stored data is read from the R AMD 13 and D 14 as the data OR 12 and OR 13 and supplied to the output data selection circuit 408.
- reading of data is performed based on address data supplied from the address selection circuit 405, as in the case of writing data.
- the interleaver 100 performs PCCC with a coding rate of “1/6 or more” by the coding device 1 and a data capacity of “16 km or less”. It is possible to perform random in-leave and delay for one symbol input data.
- the coding apparatus 1 performs SCC with a coding rate of “1/3 or more”, and that the input data capacity is “8 kilobytes or less”.
- the interleaver 100 has six RAMs D 01, D 02, D 16 among the 16 RAMs D 01, D 02,..., D 16 as shown in FIG. 60A, for example.
- D 03, D 04, D 05, D 07 are used for delay, and as shown in FIG.B, eight R AMDs 09, D 10, D 11, D 12, D 13, D 14, D 15, D 16 Used for evening leave.
- the RAM for the address as shown in Fig. C, any four of the six RAMAs may be used. Therefore, the interleaver 100 and the address storage circuit 110 do not use two RAMs D06 and D08 and two RAMAs as shown in FIG.
- the interleaver 100 includes R AMD 01, D 02, D 05, D 09, D 10, D 13, and D 14 as shown in FIGS. Are used as the puncture A (Ao) described above, and the RAMD 03, D 04, D 07, D 11, D 12, D 15, and D 16 are used as the above-described bank B (Bo). That is, when writing data to RAMD 01, D 02, D 05, D 09, D 10, D 13, and D 14, RAM D 03, D 04 , D 07, D 11, D 12, D 15, D 16 Read data from RAM D 03, D 04, D 07, D 11, D 11, D 15, D 16 When the data is written in the memory, the data is read out from RAMD01, D02, D55, D09, D10, D13, D14.
- delay data D 0 and D 1 are supplied to the R AMD 01 as input data IR 00 from the input data selection circuit 406 and written. It is. At this time, data D 0 and D 1 for 0 to 4 kilowords are written to RAMD 01.
- the RAMD 05 is supplied with delay data D 2 and D 3 as data IR 04 from the input data selection circuit 406 based on the address data AR 04 supplied from the address selection circuit 405. , Written. At this time, data D2 and D3 for 0 to 4 kilowords are written to RAMD05.
- the RAMD02 receives the delay data D4, D5 as data IR01 from the input data selection circuit 406. Is supplied and written. At this time, 0 to 4 kilowords of data D 4 and D 5 are written to RAMD 02.
- the stored data is read out from RAMD 03, D 04, D 07 as data OR 02, OR 03, OR 06 and supplied to output data selection circuit 408. Is done. Note that data reading is data writing As in the case of the above, the determination is performed based on the address data supplied from the address selection circuit 405.
- the input data selection circuit 406 receives the delay data DO, D1 is supplied and written. At this time, data D 0 and D 1 for 0 to 4 kilowords are written to RAMD 03. Also, based on the address data AR 06 supplied from the address selection circuit 405, the R AMD 07 is supplied with delay data D 2 and D 3 as data IR 06 from the input data selection circuit 406, Written. At this time, data D2 and D3 for 0 to 4 kilowords are written to RAMD07. Further, based on the address data AR03 supplied from the address selection circuit 405, the delay data D4: D5 is supplied from the input data selection circuit 406 as data IR03 to the RAMD04. Is written. At this time, data D 4 and D 5 for 0 to 4 kilowords are written in RAMD 04.
- the stored data is read out from RAMD 01, D 02, and D 05 as data OR 00, 0 R 01, OR 04, respectively, and output data selection circuit 408. Note that the data reading is performed based on the address data supplied from the address selection circuit 405 as in the data writing.
- RAMD 09, D 10, D 11, D 12, D 13, D 14, D 15, and D 16 are each based on the partial light control signal PW, It functions as a RAM, and acts as a RAM having a storage capacity of 8 bits x 8192 words in a pseudo manner.
- the R AMD 13 is supplied with the data I 0 for the overnight read as data IR 12 from the input data selection circuit 406 and is written. It is. At this time, data I 0 for 0 to 8 kilowords is written to RAMD 13. Also, in the RAMD 14, similarly to the RAMD 13, based on the address data AR 13 supplied from the address selection circuit 405, the data IR 13 is provided from the input data selection circuit 406. Then, data I 0 for interleaving is supplied and written. At this time, the data I 0 for 0 to 8 kilowords is written in the RAM 14.
- R AMD 09 receives an input data selection circuit 406 from the address selection circuit 406 based on the address data AR 08 supplied from the address selection circuit 405 and outputs a data IR 08 to the interleave.
- Data I1 is provided and written. At this time, data I1 for 0 to 8 kilowords is written to RAMD09.
- the input data overnight selection circuit 406 receives data IR 09 as data IR 09.
- Data I1 for the interleave is supplied and written. At this time, data I 1 for 0 to 8 kilowords is written to RAMD 10.
- the stored data is read out as data OR 10 and 0 R 14, respectively.
- the data is supplied to the output data selection circuit 408 as data.
- the stored data is read out as data OR 11 and OR 15, respectively, and the other one symbol data out of the two symbol data is read. As an overnight, it is supplied to the output data overnight selection circuit 408. Note that reading of data is performed based on address data supplied from the address selection circuit 405, as in the case of writing data.
- the R AMD 15 receives the input data from the input data selection circuit 406 as the data IR 14, and outputs the data for interleaving.
- the data I 0 is supplied and written.
- data I 0 for 0 to 8 kilowords is written to RAMD 15.
- RAMD 16 receives data IR 15 from input data selection circuit 40 6 based on address data AR 15 supplied from address selection circuit 400.
- data 10 for in-leave is supplied and written.
- data I 0 for 0 to 8 kilowords is written to R AMD 16.
- the R AMD 11 receives an interface from the input data selection circuit 406 as a data IR 10 based on the address data AR 10 supplied from the address selection circuit 405. Data I1 for one leave is supplied and written. At this time, RAMD 11 is written with 0 to 8 kilowords of data I 1. Similarly to RAMD 11, RAMD 12 is interleaved from input data selection circuit 406 as data IR 11 based on address data AR 11 supplied from address selection circuit 405. Is supplied and written. At this time, data I 1 for 0 to 8 kilowords is written to R AMD 12.
- the stored data is read out as data OR 08, 0 R 12, and one of the two symbol data
- the data is supplied to the output data selection circuit 408 as a symbol data.
- the stored data is read out as data OR 09 and 0 R 13, respectively.
- the other one symbol data is Is supplied to the output data selection circuit 408. Note that reading of data is performed based on address data supplied from the address selection circuit 405, as in the case of writing data.
- the interleaver 100 performs SCCC with a coding rate of "1/3 or more" by the coding device 1, and the capacity of the data overnight is "8 kilowords or less". Random interleaving and delay can be applied to the input data of the two symbols.
- the interleaver 100 it is necessary for the interleaver 100 to interleave the data of two symbols and to delay the data of four symbols.
- the interleaver 100 is, for example, as shown in FIG. 61A, out of the 16 RAMs D 01, D 02,..., D 16, eight RAMs D 01, D 02, D 03, D 04, D 05, D 06, D 07, D 08 are used for delay, and eight RAMs D 09, D 10, D 11, D 12, D 13, D 14, D 15, and D 16 are used for in-leave.
- the RAM for the address as shown in Fig. C, all six RAMAs are used.
- the interleaver 100 includes R AMD 01, D 02, D 05, D 06, D 09, D 10, D 13, as shown in FIGS. D14 is used as the bank A ( ⁇ ⁇ ⁇ ., ⁇ ,) described above, and RAMD03, D04, D07, D08, D11, D12, D15, and D16 are used as the bank A described above.
- B (Used as B., B.
- the in-lever-reaver 100 is designed for RAMD01, D02, D05: D06, D09, D10, D13, D14.
- the input data selection circuit 406 Based on the address data AR 00, AR 01 supplied from the address selection circuit 405, the input data selection circuit 406 outputs data R 00, IR 0 1 to R AMD 01, DO 2. As a result, delay data D 0 and D 1 are supplied and written. At this time, the RAMD 02 stores data D0 and D1 only in a half storage area in the word direction as shown by the hatched portion in FIG. never. That is, data of 0 to 4 kilowords of the data D O and D 1 is written to RAMD 01, and data of 4 to 6 kilowords is written to RAMD 02.
- the RAMD 05 and DO 6 are provided with data IR 04 and IR 05 from the input data selection circuit 406 based on the address data AR 04 and AR 05 supplied from the address selection circuit 405, respectively. As a result, delay data D 2 and D 3 are supplied and written. At this time, the RAM DO 6 stores data D 2 and D 3 only in half the memory area in the word direction, as shown in the hatched area in FIG. never remembers the night. That is, of the data D 2 and D 3, data of ⁇ to 4 kilowords are written to the RAMD 05, and data of 4 to 6 kilowords are written to the RAMD 06.
- RAMD 03, D 04, D 07, and D 08 the stored data is read as data OR 02, 0 R 03, OR 06, OR 07. And output to the output data selection circuit 408.
- RAMD 04 and DO 8 each store data only in half the storage area in the word direction, as indicated by the hatched area in FIG. Evening is not remembered. Note that the data reading is performed based on the address data supplied from the address selection circuit 405 in the same manner as in the data writing.
- RAMD 03 and DO 4 respectively receive the data IR 02 and IR 03 from the input data selection circuit 406.
- the delay data D 0 and D 1 are supplied and written.
- RAMD 04 stores data D 0 and D 1 only in a half storage area in the word direction, as shown by the hatched portion in FIG. A, and does not store data in the remaining storage areas. Absent. That is, data of 0 to 4 kilowords of the data D O: D1 is written to the RAMD 03, and data of 4 to 6 kilowords is written to the RAMD 04.
- the RAMD 07 and D 08 receive the data IR 06 and IR 07 from the input data selection circuit 406, respectively.
- Data D 2 and D 3 for delay are supplied and written.
- RAMD 08 stores data D 2 and D 3 only in half the memory area in the word direction, as shown in the hatched area in FIG. never remembers the evening. That is, 0 to 4 kilowords of data D2 and D3 are written to RAMD07, and 4 to 6 kilowords of data are written to RAMD08.
- RAMD 01, D 02, D 05, and D 06 the stored data can be stored as data OR 00, 0 R 0 1, 0 R 04, OR 05
- the data is read and supplied to the output data selection circuit 408.
- RAMD 02 and DO 6 each store the data only in half the storage area in the word direction, as indicated by the shaded area in FIG. , The night is not remembered. Note that the data reading is performed based on the address data supplied from the address selection circuit 405, as in the data writing.
- RAMD 09, D 10, D 11, D 12, D 13, D 14, D 15, D 16 functions as a partial-write RAM based on the partial-write control signal PW, and acts as a RAM having a storage capacity of 8 bits ⁇ 8192 words in a pseudo manner.
- the input data selecting circuit 406 Based on the address data AR 12 and AR 13 supplied from the address selecting circuit 405, the input data selecting circuit 406 outputs the data IR 12 to R AMD 13 and D 14 respectively. , IR 13, interleaving data I 0 is supplied and written. At this time, the RAMD 14 stores data I0 only in half the storage area in the card direction, and stores data in the remaining storage area, as indicated by the shaded area in FIG. There is no. That is, data of 0 to 8 kilowords out of the data I 0 is written to R AMD 13, and data of 8 to 12 kilowords is written to RAMD 14.
- RAMD 10 stores data I 1 only in half the memory area in the word direction, as shown in the hatched area in FIG. Is not remembered. That is, data of 0 to 8 kilowords of data I1 is written to RAMD09, and data of 8 to 12 kilowords is written to RAMD10.
- the stored data is read out from R AMD 15 and D 16 as data OR 14 and OR 15, respectively. In the evening, it is supplied to the output data selection circuit 408.
- the RAMD 16 stores data only in a half storage area in the word direction, as indicated by the hatched portion in FIG. B, and no data is stored in the remaining storage areas.
- the stored data is read out as data OR 10 and OR 11, respectively, and one symbol data of the other one of the two symbol data Is supplied to the output data selection circuit 408.
- the RAMD 12 stores data only in half the storage area in the pad direction, as in the RAMD 16, and the remaining storage area , No data is stored. Note that reading of data is performed based on address data supplied from the address selection circuit 405, as in the case of writing data.
- the input data selection circuit 406 transmits the data IR 1 to R AMD 15 and D 16 respectively.
- Interleave data I0 is supplied and written as IR15.
- the RAMD 16 stores the data I0 only in the half storage area in the word direction as shown by the hatched portion in FIG. B, and does not store the data in the remaining storage areas. That is, data of 0 to 8 kilowords out of the data I 0 is written to R AMD 15, and data of 8 to 12 kilowords are written to RAMD 16.
- R AMD 11 and D 12 each have a data IR 1 from the input data selection circuit 406 based on the address data AR 10 and AR 11 supplied from the address selection circuit 405. 0, IR11
- the interleave data I1 is supplied and written.
- R AMD 12 stores data I 1 only in half the storage area in the word direction, as in RAMD 16, and stores the remaining data in the remaining storage area. No data is stored. That is, data of 0 to 8 kilowords out of data I1 is written to RAMD11, and data of 8 to 12 kilowords is written to RAMD12.
- R AMD 13 and D 14 the stored data is read as data OR 12 and 0 R 13, respectively, and one of the two symbol data The data is supplied to the output data selection circuit 408 as a symbol.
- the RAMD 14 stores the data only in a half storage area in the word direction, as indicated by the hatched portion in FIG. B, and no data is stored in the remaining storage areas.
- R AMD 09, D 10 the stored data is read out as data OR 08, 0 R 09, and the symbol data of the other one of the two symbol data is read out. In the evening, it is supplied to the output data selection circuit 408. At this time, as shown by the hatched portion in FIG.
- RAMD 10 stores the data only in half of the memory area in the memory card direction in the same manner as RAMD 14, and the remaining memory area In No data is stored. Note that data is read based on the address data supplied from the address selection circuit 405 as in the data writing.
- the interleaver 100 performs the punctured SCCC by the encoder 1 and the input data of the two symbols whose data capacity is “less than or equal to 12 kilowords”. May be subject to in-line in-between leave and delay.
- the interleaver 100 has eight RAMD 01, D 02 among 16 RAMD 01, D 02,..., D 16 as shown in FIG. 62A, for example. , D 03, D 04, D 05, D 06, D 07 D 08 are used as delays, and as shown in FIG. B, eight RAMs D 09, D 10, D ll, D 12, D 12 13. Use D3, D14, D15, and D16 for in-leaving.
- the RAM for the address as shown in Fig. C, any four of the six RAMAs may be used. Therefore, as shown in FIG. D, the in-all-in-one-reaver 100 and the address storage circuit 110 do not use two RAMAs.
- the interleaver 100 includes R AMD 01, D 02, D 05, D 06, D 09, D 10, and D 13 as shown in FIGS. , D14 are used as the above-mentioned bank A (A., Ai), and RAMD03, D04, D07, D08, Dll, D12, D15, and D16 are used as the above-mentioned bank A.
- B used as B (B., ⁇ ,). That is, the input / output unit 100 writes data to RAMD 01, D 02, D 05: D 06, D 09, D 10, D 13, and D 14.
- RAMD 13, D 14 and RAMD 09, D 10 operate based on the same address
- the input data selection circuit 406 Based on the address data AR 00 and AR 01 supplied from the address selection circuit 405, the input data selection circuit 406 outputs R AMD 01 and D 02 as data IR 00 and IR 01, respectively.
- Delay data D 0 and D 1 are supplied and written. At this time, data of 0 to 4 kilowords of data D0 and D1 is written to RAMD01, and data of 4 to 8 kilowords is written to RAMD02. Further, the RAMD 05 and DO 6 are provided with data IR 04 and IR 05 from the input data selection circuit 406 based on the address data AR 04 and AR 05 supplied from the address selection circuit 405, respectively.
- delay data D2 and D3 are supplied and written. At this time, data of 0 to 4 kilowords of the data D2 and D3 is written to RAMD05, and data of 4 to 8 kilowords is written to RAMD06.
- the stored data is read out as data OR 02, 0 R 03, 0 R 06, OR 07
- the data is supplied to the output data selection circuit 408. Note that data is read based on the address data supplied from the address selection circuit 405 in the same manner as in the data write operation.
- the RAMD 03 and DO 4 are provided with data IR 02 and IR 03 from the input data selection circuit 406 based on the address data AR 02 and AR 03 supplied from the address selection circuit 405, respectively.
- the delay data D 0 and D 1 are supplied and written.
- data of 0 to 4 kilowords of data D0 and D1 is written to RAMD03
- data of 4 to 8 kilowords is written to RAMD04.
- the RAMD 07 and DO 8 are provided with data IR 06 and IR 07 from the input data selection circuit 406 based on the address data AR 06 and AR 07 supplied from the address selection circuit 405, respectively.
- the delay data D 2 and D 3 are supplied and written.
- RAMD 07 contains Data of 0 to 4 kilowords of data D2 and D3 are written, and data of 4 to 8 kilobytes are written to RAM DO8.
- the stored data is read as data OR00, OR01, 0R04, OR05. And supplied to the output data selection circuit 408. Note that reading of data is performed based on address data supplied from the address selection circuit 405, as in the case of writing data.
- RAMD 09, D 10, D 11, D 12, D 13, D 14, D 15, and D 16 each have a partial light control signal PW based on the partial light control signal PW. It functions as a RAM with a storage capacity of 8 bits ⁇ 8192 words in a pseudo manner.
- the input data selection circuit 406 Based on the address data AR 12 and AR 13 supplied from the address selection circuit 405, the input data selection circuit 406 outputs the data IR 1 2 and R 1
- the data I 0 for interleaving is supplied and written as IR 13.
- data of 0 to 8 kilowords out of the data 10 is written to the RAMD 13, and data of 8 to 16 kilowords is written to the RAMD 14.
- R AMD 09 and D 10 are provided as data IR 08 and IR 09 from the input data overnight selection circuit 406 based on the address data AR 08 and AR 09 supplied from the address selection circuit 405.
- the data I1 for the in-live recording is supplied and written. At this time, data of 0 to 8 kilowords of the data I-I are written to RAMD09, and data of 8 to 16 kilowords are written to RAMD10.
- the stored data is read out as data OR 14 and 0 R 15, respectively.
- the data is supplied to the output data selection circuit 408 as symbol data.
- R AMD 11 and D 12 the stored data is read out as data OR 10 and 0 R 11, respectively, and the other one of the two symbol data is used as the data of the other symbol.
- the data is supplied to the output data selection circuit 408 as volt data. Note that data is read from the address selection circuit 405 in the same way as when data is written. It will be performed based on Rudoles De Night.
- the data IR 14 and the data IR 14 are supplied to the R AMD 15 and D 16 respectively.
- the data 10 for the in-leave is supplied and written.
- data of 0 to 8 kilowords out of the data 10 is written to the RAMD 15 and data of 8 to 16 kilowords is written to the RAMD 16.
- R AMD 11 and D 12 each receive data IR from input data overnight selection circuit 406 based on address data AR 10 and AR 11 supplied from address selection circuit 405.
- 10 and IR 11 are supplied and written as data I 1 for in-leave. At this time, data of 0 to 8 kilowords of the data I 1 is written to the RAMD 11, and data of 8 to 16 kilowords is written to the RAM D 12.
- the stored data is read out from R AMD 13 and D 14 as data OR 12 and OR 13, respectively. In the evening, it is supplied to the output data selection circuit 408. From R AMD 09 and D 10, the stored data is read out as data OR 08 and 0 R 09, respectively. The data is supplied to the output data selection circuit 408 as port data. The reading of the data is performed based on the address data supplied from the address selection circuit 405 as in the data writing.
- the interleaver 100 can perform pairwise interleaving and delay on the input data of the two symbols on which the SCC is performed by the encoder 1.
- the coding apparatus 1 performs SCCC with a coding rate of “1/3 or more” and that the capacity of input data is “4 kilowords or less”.
- the interleaver 100 it is necessary for the interleaver 100 to interleave data of three symbols and to interleave data of four symbols.
- the interleaver 100 has four RAMD 0 1, D 03, D 16 out of 16 RAMs D 01, D 02,..., D 16. 05 and D 07 are used as delays, and as shown in FIG. B, 12 RAMs D 02, D 04, D 06, D 08, D 09, D 10, D 11, D 12, D 12 13, D 14, D 15, D 16 are used for the interleave.
- the RAM for the address as shown in FIG. C, any three of the six RAMAs may be used. Therefore, the interleaver 100 and the address storage circuit 110 do not use three RAMAs as shown in FIG.
- the Inuichi Reaver 100 has R AMD 01, D 02, D 05, D 06, D 09, D 10, D 13, Using D 14 as the above-described bank A (Ao), RAMD 03, D 04, D 07, D 08, D 11, D 12, D 15, D 16 are used as the above-described bank B (Bo) Used as In other words, if the data is written to RAMD01, D02, D05, D06, D09, D10, D13, D14 , RA MD 03, D 04, D 07, D 08, D 11, D 12, D 15, D 16, read data from RAMD 03, D 04, D 07, D 08, D ll, When data is written to D12, D15, D16, RAMD01, DO2, D05, D06, D09, D10, D13, D1 Read data from 4.
- R AMD 01 Based on the address data AR 00 supplied from the address selection circuit 405, R AMD 01 has delay data D 0 and D 1 as input data IR 00 from the input data selection circuit 406. Supplied and written. At this time, RAMD 01 stores data D 0 and D 1 only in half of the memory area in the direction of the arrow, as indicated by the shaded area in FIG. I do not remember. That is, data D 0 and D 1 for 0 to 2 kilowords are written to RAMD 01. Further, based on the address data AR04 supplied from the address selection circuit 405, the delay data D2 and D3 are supplied from the input data selection circuit 406 as data IR04 to the RAMD05. Is written.
Landscapes
- Physics & Mathematics (AREA)
- Probability & Statistics with Applications (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Error Detection And Correction (AREA)
- Detection And Correction Of Errors (AREA)
Description
Claims
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| EP01961290A EP1315302A4 (en) | 2000-08-31 | 2001-08-31 | SOFT-OUTPUT DECODER |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2000-263120 | 2000-08-31 | ||
| JP2000263120A JP2002076933A (ja) | 2000-08-31 | 2000-08-31 | 軟出力復号装置及び軟出力復号方法、並びに、復号装置及び復号方法 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| WO2002019539A1 WO2002019539A1 (fr) | 2002-03-07 |
| WO2002019539A9 true WO2002019539A9 (fr) | 2002-11-28 |
Family
ID=18750708
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/JP2001/007576 Ceased WO2002019539A1 (fr) | 2000-08-31 | 2001-08-31 | Decodeur a sortie ponderee |
Country Status (4)
| Country | Link |
|---|---|
| US (1) | US7180968B2 (ja) |
| EP (1) | EP1315302A4 (ja) |
| JP (1) | JP2002076933A (ja) |
| WO (1) | WO2002019539A1 (ja) |
Families Citing this family (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003174650A (ja) * | 2001-09-27 | 2003-06-20 | Canon Inc | 画像符号化装置、可変長符号化装置、制御装置およびそれらの方法 |
| US7137059B2 (en) * | 2002-11-20 | 2006-11-14 | Broadcom Corporation | Single stage implementation of min*, max*, min and /or max to perform state metric calculation in SISO decoder |
| US7089481B2 (en) * | 2002-07-22 | 2006-08-08 | Agere Systems Inc. | High speed arithmetic operations for use in turbo decoders |
| GB0229320D0 (en) * | 2002-12-17 | 2003-01-22 | Koninkl Philips Electronics Nv | Signal processing method and apparatus |
| DE602004018489D1 (de) * | 2003-05-16 | 2009-01-29 | Thomson Licensing | Demodulierung und wiederholungsdecodierung von mehrschichtigen signalen |
| JP4320661B2 (ja) * | 2003-06-30 | 2009-08-26 | 三菱電機株式会社 | 光受信装置および光受信方法 |
| TWI228654B (en) * | 2003-07-11 | 2005-03-01 | Mediatek Inc | Non-binary Viterbi data processing system and method |
| KR100723876B1 (ko) * | 2005-05-12 | 2007-05-31 | 한국전자통신연구원 | 멀티레벨변조신호 기반의 검파 장치, 반복 수신 장치 및이를 이용한 방법 |
| US7395461B2 (en) * | 2005-05-18 | 2008-07-01 | Seagate Technology Llc | Low complexity pseudo-random interleaver |
| US20070011557A1 (en) * | 2005-07-07 | 2007-01-11 | Highdimension Ltd. | Inter-sequence permutation turbo code system and operation methods thereof |
| US7797615B2 (en) * | 2005-07-07 | 2010-09-14 | Acer Incorporated | Utilizing variable-length inputs in an inter-sequence permutation turbo code system |
| US7856579B2 (en) * | 2006-04-28 | 2010-12-21 | Industrial Technology Research Institute | Network for permutation or de-permutation utilized by channel coding algorithm |
| US7386823B2 (en) * | 2005-07-20 | 2008-06-10 | Springsoft, Inc. | Rule-based schematic diagram generator |
| US7526715B2 (en) * | 2005-10-17 | 2009-04-28 | Ramot At Tel Aviv University Ltd. | Probabilistic error correction in multi-bit-per-cell flash memory |
| US7590925B2 (en) * | 2006-09-05 | 2009-09-15 | Mediatek Inc. | Apparatus and method for detecting puncture position in a symbol stream encoded by punctured convolutional coding scheme |
| US20080320365A1 (en) * | 2007-06-20 | 2008-12-25 | Texas Instruments Incorporated | Providing an initial syndrome to a crc next-state decoder independently of its syndrome feedback loop |
| US8271861B2 (en) * | 2008-05-09 | 2012-09-18 | Marvell International Ltd. | Symbol vector-level combining receiver for incremental redundancy HARQ with MIMO |
| US10022468B2 (en) * | 2009-02-02 | 2018-07-17 | Kimberly-Clark Worldwide, Inc. | Absorbent articles containing a multifunctional gel |
| US8516333B1 (en) * | 2009-11-19 | 2013-08-20 | Viasat, Inc. | Pre-interleaving for forward error correction codes |
| JP6632876B2 (ja) * | 2015-12-04 | 2020-01-22 | シナプティクス・ジャパン合同会社 | バッファメモリ装置及び表示駆動デバイス |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0740669B2 (ja) * | 1986-04-16 | 1995-05-01 | 株式会社日立製作所 | 最尤復号器 |
| JPH052842A (ja) * | 1991-06-25 | 1993-01-08 | Sony Corp | 軟判定復号回路 |
| US5983384A (en) | 1997-04-21 | 1999-11-09 | General Electric Company | Turbo-coding with staged data transmission and processing |
| JP2000068862A (ja) | 1998-08-19 | 2000-03-03 | Fujitsu Ltd | 誤り訂正符号化装置 |
| FR2805106A1 (fr) * | 2000-02-14 | 2001-08-17 | Mitsubishi Electric Inf Tech | Procede de transmission numerique de type a codage correcteur d'erreurs |
| JP2001285084A (ja) * | 2000-03-31 | 2001-10-12 | Sony Corp | 符号化装置、符号化方法及び符号化プログラムが記録された記録媒体、並びに、復号装置、復号方法及び復号プログラムが記録された記録媒体 |
-
2000
- 2000-08-31 JP JP2000263120A patent/JP2002076933A/ja not_active Withdrawn
-
2001
- 2001-08-31 WO PCT/JP2001/007576 patent/WO2002019539A1/ja not_active Ceased
- 2001-08-31 US US10/111,724 patent/US7180968B2/en not_active Expired - Fee Related
- 2001-08-31 EP EP01961290A patent/EP1315302A4/en not_active Withdrawn
Also Published As
| Publication number | Publication date |
|---|---|
| WO2002019539A1 (fr) | 2002-03-07 |
| US20030061003A1 (en) | 2003-03-27 |
| EP1315302A4 (en) | 2007-05-23 |
| US7180968B2 (en) | 2007-02-20 |
| EP1315302A1 (en) | 2003-05-28 |
| JP2002076933A (ja) | 2002-03-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2002019539A9 (fr) | Decodeur a sortie ponderee | |
| JP2002076915A (ja) | インターリーブ装置及びインターリーブ方法、並びに、復号装置及び復号方法 | |
| US7715503B2 (en) | Parallel concatenated code with soft-in soft-out interactive turbo decoder | |
| JP2002076925A (ja) | 軟出力復号装置及び軟出力復号方法、並びに、復号装置及び復号方法 | |
| JP4427883B2 (ja) | インターリーブ装置及びインターリーブ方法、並びに、復号装置及び復号方法 | |
| US7225306B2 (en) | Efficient address generation for Forney's modular periodic interleavers | |
| KR100416569B1 (ko) | 터보 치환기 및 이를 이용한 터보 복호기 | |
| JP4543522B2 (ja) | 軟出力復号装置及び軟出力復号方法、並びに、復号装置及び復号方法 | |
| JP4380041B2 (ja) | 軟出力復号装置及び軟出力復号方法、並びに、復号装置及び復号方法 | |
| JP2002076917A (ja) | インターリーブ装置及びインターリーブ方法、並びに、復号装置及び復号方法 | |
| JP2002076940A (ja) | 軟出力復号装置及び軟出力復号方法、並びに、復号装置及び復号方法 | |
| JP2002076930A (ja) | 軟出力復号装置及び軟出力復号方法、並びに、復号装置及び復号方法 | |
| JP2002076929A (ja) | 軟出力復号装置及び軟出力復号方法、並びに、復号装置及び復号方法 | |
| JP2002076914A (ja) | インターリーブ装置及びインターリーブ方法、並びに、復号装置及び復号方法 | |
| JP2002076916A (ja) | インターリーブ装置及びインターリーブ方法、並びに、復号装置及び復号方法 | |
| JP2002076934A (ja) | 軟出力復号装置及び軟出力復号方法、並びに、復号装置及び復号方法 | |
| JP2002076932A (ja) | 復号装置及び復号方法 | |
| JP2002076944A (ja) | 軟出力復号装置及び軟出力復号方法、並びに、復号装置及び復号方法 | |
| JP2002076927A (ja) | 軟出力復号装置及び軟出力復号方法、並びに、復号装置及び復号方法 | |
| JP2002077282A (ja) | 軟出力復号装置及び軟出力復号方法、並びに、復号装置及び復号方法 | |
| JP2002076938A (ja) | 軟出力復号装置及び軟出力復号方法、並びに、復号装置及び復号方法 | |
| JP2002076941A (ja) | 軟出力復号装置及び軟出力復号方法、並びに、復号装置及び復号方法 | |
| JP2002076937A (ja) | 軟出力復号装置及び軟出力復号方法、並びに、復号装置及び復号方法 | |
| JP2002076936A (ja) | 軟出力復号装置及び軟出力復号方法、並びに、復号装置及び復号方法 | |
| JP2002076928A (ja) | 復号装置 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| AK | Designated states |
Kind code of ref document: A1 Designated state(s): US |
|
| AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE TR |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 10111724 Country of ref document: US |
|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
| WWE | Wipo information: entry into national phase |
Ref document number: 2001961290 Country of ref document: EP |
|
| AK | Designated states |
Kind code of ref document: C2 Designated state(s): US |
|
| AL | Designated countries for regional patents |
Kind code of ref document: C2 Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE TR |
|
| WWP | Wipo information: published in national office |
Ref document number: 2001961290 Country of ref document: EP |
|
| WWW | Wipo information: withdrawn in national office |
Ref document number: 2001961290 Country of ref document: EP |