WO2019093855A1 - Harq를 지원하기 위해 폴라 코드를 기반으로 부호화하는 방법 및 이를 이용한 무선 장치 - Google Patents
Harq를 지원하기 위해 폴라 코드를 기반으로 부호화하는 방법 및 이를 이용한 무선 장치 Download PDFInfo
- Publication number
- WO2019093855A1 WO2019093855A1 PCT/KR2018/013793 KR2018013793W WO2019093855A1 WO 2019093855 A1 WO2019093855 A1 WO 2019093855A1 KR 2018013793 W KR2018013793 W KR 2018013793W WO 2019093855 A1 WO2019093855 A1 WO 2019093855A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- input vector
- information
- index
- bit
- mother matrix
- 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
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/12—Arrangements for detecting or preventing errors in the information received by using return channel
- H04L1/16—Arrangements for detecting or preventing errors in the information received by using return channel in which the return channel carries supervisory signals, e.g. repetition request signals
- H04L1/18—Automatic repetition systems, e.g. Van Duuren systems
- H04L1/1812—Hybrid protocols; Hybrid automatic repeat request [HARQ]
-
- 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
- H03M13/63—Joint error correction and other techniques
- H03M13/6306—Error control coding in combination with Automatic Repeat reQuest [ARQ] and diversity transmission, e.g. coding schemes for the multiple transmission of the same information or the transmission of incremental redundancy
-
- 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
- H03M13/03—Error detection or forward error correction by redundancy in data representation, i.e. code words containing more digits than the source words
- H03M13/05—Error detection or forward error correction by redundancy in data representation, i.e. code words containing more digits than the source words using block codes, i.e. a predetermined number of check bits joined to a predetermined number of information bits
- H03M13/13—Linear 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
- H03M13/63—Joint error correction and other techniques
- H03M13/635—Error control coding in combination with rate matching
- H03M13/6362—Error control coding in combination with rate matching by puncturing
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/004—Arrangements for detecting or preventing errors in the information received by using forward error control
- H04L1/0041—Arrangements at the transmitter end
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/004—Arrangements for detecting or preventing errors in the information received by using forward error control
- H04L1/0056—Systems characterized by the type of code used
- H04L1/0057—Block codes
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/004—Arrangements for detecting or preventing errors in the information received by using forward error control
- H04L1/0056—Systems characterized by the type of code used
- H04L1/0067—Rate matching
- H04L1/0068—Rate matching by puncturing
- H04L1/0069—Puncturing patterns
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/004—Arrangements for detecting or preventing errors in the information received by using forward error control
- H04L1/0056—Systems characterized by the type of code used
- H04L1/0071—Use of interleaving
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/12—Arrangements for detecting or preventing errors in the information received by using return channel
- H04L1/16—Arrangements for detecting or preventing errors in the information received by using return channel in which the return channel carries supervisory signals, e.g. repetition request signals
- H04L1/18—Automatic repetition systems, e.g. Van Duuren systems
- H04L1/1812—Hybrid protocols; Hybrid automatic repeat request [HARQ]
- H04L1/1819—Hybrid protocols; Hybrid automatic repeat request [HARQ] with retransmission of additional or different redundancy
Definitions
- the present invention relates to wireless communication, and more particularly, to a method of encoding based on polar codes to support HARQ and a wireless device using the same.
- HARQ is a technique for recovering an error by requesting retransmission when a packet with an error is received.
- a method for coding based on polar codes to support HARQ is characterized in that a wireless device constructs an input vector of length N for retransmission applied to a predetermined mother matrix based on polar codes, A second index set for the punctured position among the encoded bits generated based on the mother matrix, and a second set of indexes for at least one unoccupied frozen bit associated with the information copy technique A third set of indexes for the location; And the wireless device performing encoding based on the input vector and the mother matrix.
- 1 is a diagram showing the concept of polar encoding.
- Fig. 2 is a diagram showing polarization tendency according to the length of a code block.
- FIG. 3 is a diagram showing a base module of a polar code.
- FIG. 4 is a diagram of an example of a channel combination having N levels.
- FIG. 5 shows a conceptual diagram of reciprocal characteristics according to the perforation according to the present embodiment.
- FIG. 6 is a conceptual diagram of a puncturing pattern based on a hierarchical characteristic according to the puncturing according to the present embodiment.
- FIG. 7 is a conceptual diagram of a method of encoding based on polar codes to support HARQ according to the present embodiment.
- FIG. 8 is a flowchart illustrating a method of encoding based on polar codes to support HARQ according to the present embodiment.
- FIG. 9 is a flowchart illustrating a procedure for performing encoding based on polar codes to support HARQ according to the present embodiment.
- FIG. 10 is a flowchart illustrating a procedure for decoding based on polar codes to support HARQ according to an embodiment of the present invention.
- 11 is a conceptual diagram showing the characteristics of a permuted puncturing pattern of successive puncturing according to the present embodiment.
- FIGS. 12 and 13 are conceptual diagrams illustrating a method of encoding based on a polar code in order to support HARQ according to another embodiment of the present invention.
- FIG. 14 shows a simulation result of a polar code designing a variable code rate according to the present embodiment.
- 15 is a diagram for explaining wireless devices to which the present invention examples can be applied.
- 1 is a diagram showing the concept of polar encoding.
- the polar encoding process using polar codes may include a channel combining step S1 and a channel splitting step S2.
- a vector channel Wvcc may be generated by appropriately combining a plurality of basic channels W.
- a channel having a channel capacity C (W) of '1' can be allocated for transmission of an information bit because error-free transmission is possible.
- a channel having a channel capacity C (W) of '0' can be allocated for transmission of a frozen bit which is a meaningless bit because information can not be transmitted.
- Fig. 2 is a diagram showing polarization tendency according to the length of a code block.
- FIG. 3 is a diagram showing a base module of a polar code.
- the channel combination described above may be performed by parallel permutation of a B-DMC channel (binary-input discrete memoryless channel).
- the size of the code block may be determined based on the channel combination.
- W in FIG. 3 may represent a B-DMC channel.
- the base module of the polar code of FIG. 3 can combine two B-DMC channels.
- u1 and u2 in FIG. 3 may be a binary-input source bit.
- X1 and x2 in FIG. 3 may be inputs to the corresponding channel W.
- y1 and y2 may be output coded bits.
- W2 a channel equivalent to the entire channel
- x1 may include the result of the XOR operation of u1 and u2.
- u2 may be included in x2 without additional computation.
- Equation 1 the basic matrix F for the basic module of the polar code of FIG. 3 can be expressed as Equation 1 below.
- each channel can be represented in a recursive form.
- the x vector of the output bit of the encoder can be expressed using the following equation (2).
- Equation (3) the x vector of the output bit of the encoder is expressed by the following Equation (3), and the u vector of the input bit of the encoder can be expressed by Equation (4) below.
- the code rate is N / K.
- the K bits can be understood as bits for the payload.
- the generator matrix GN of Equation (5) can be expressed in the form of a Kronecker product of the base matrix (F) of Equation (1).
- a generator matrix G1 of size 2X2 may correspond to Equation (1).
- a generator matrix G2 of size 4X4 may be as shown in Table 1 below.
- the generator matrix G3 of size 8X8 according to Equations (5) and (6) may be as shown in Table 2 below.
- the generation matrix GN according to Equation (5) above may be referred to as a predetermined mother matrix based on polar codes.
- FIG. 4 is a diagram of an example of a channel combination having N levels.
- FIG. 4 can be understood as a case where an operation for Bit-reverse permutation exists in the encoding part.
- the size (N) of the code block may have a limitation of 2? N (n is a natural number).
- the decoding of the polar code may be performed using successive cancellation (SC) decoding.
- the continuous cancellation decoding method may be performed by calculating a transition probability of a channel, and then calculating a likelihood ratio (LLR) of an input bit based on the calculated probability.
- LLR likelihood ratio
- the channel transition probability can be calculated in a recursive form using the fact that the channel combination and the channel segmentation process are recursive. That is, the likelihood ratio value can be calculated in a recursive form.
- the payload bit to which the data is transmitted is a data bit
- the remaining bits except for the payload bit i.e., the bit where the actual data is not transmitted
- the bits are determined as data bits or frozen bits.
- the location of the data bits and the location of the frozen bits may be determined based on the channel capacity of each channel.
- FIG. 5 shows a conceptual diagram of reciprocal characteristics according to the perforation according to the present embodiment.
- the generation matrix GN i.e., the input vector to be applied to the mother matrix
- m is a natural number ⁇ .
- the input vector for the generator matrix G3 of size 8X8 in Table 2 is ⁇ u0, u1, ... , u7 ⁇ .
- the position index for the input vector is ⁇ 0,1, ... N-1 ⁇ .
- an input vector of length 8 (e.g., u0, ..., u7) may be encoded into a vector set (e.g., x0, ... x7) encoded based on a polar code of length 8 according to equation .
- PN puncturing pattern
- an index of punctured bits (e.g., y0, y1) (for example, , y2y3 may be the same.
- input vectors corresponding to punctured bits e.g., y0, y1, y2y3 (e.g., u0, u1, u2, u3) It can be set as a vector set.
- FIG. 6 is a conceptual diagram of a puncturing pattern based on a hierarchical characteristic according to the puncturing according to the present embodiment.
- the continuous puncturing pattern when continuous puncturing is performed successively from the position index ⁇ 0 ⁇ to ⁇ 3 ⁇ , the continuous puncturing pattern can be understood as a puncturing pattern reflecting the layer characteristic.
- each input vector e.g., u0, u1, u2, u3
- B_PN ⁇ 0,1,2,3 ⁇
- the index of the information bit set Ac1 with respect to the optimized position on the polar code is ⁇ 5, 6, 7 ⁇ .
- the index of the information bit set Ac1 may be determined in consideration of the sequence and / or puncturing pattern used for the encoder input of the polar code.
- an unknown bit 'a' may be located at u5
- an unoborn bit 'b' may be located at u6
- an unannounced bit 'c' may be located at u7.
- a seed puncturing pattern can be referred to herein.
- the bit index set of the seed puncturing pattern can be defined by the following equation (8).
- a change in the relatively positioned input value can not affect the output value for the relatively positioned input value.
- the encoded bits (for example, x4, x5, x6, and x7) to be transmitted can be decoded using polar codes having a size smaller than the polar code size before puncturing.
- the transmitted encoded bits e.g., x4, x5, x6, x7 of FIG. 6 may be replaced by a length 4 mother matrix (i.e., Table 1) can be used.
- the code rate R1, K / N
- the code rate may be 3/4.
- FIG. 7 is a conceptual diagram of a method of encoding based on polar codes to support HARQ according to the present embodiment.
- the input vector for the 8X8-sized mother matrix G3 of Table 2 is ⁇ u0, u1, ..., , u7 ⁇ . That is, the position index for the input vector of FIG. 7 is ⁇ 0,1, ... ⁇ . ≪ / RTI >
- the order in which the output bit sets (e.g., y0, ... y7) are decoded can be predetermined.
- the output values i.e., y0, y4, y2, y6, y1, y5, y3, y7 encoded in the order of u0, u4, u2, u6, u1, u5, u3, u7 are sequentially decoded .
- FIG. 7 is described assuming a HARQ retransmission case after failure of a previous transmission corresponding to FIG. That is, the input vector ⁇ u0, u1, ..., , u7 ⁇ may be configured for HARQ retransmission.
- the size of the mother matrix for the polar code of FIG. 7 is the same '8' as in FIG. 6, and the codeword size of FIG. 7 may be '6' extended from the previous transmission considering HARQ retransmission.
- the size of the information bits can be assumed to be the same as the previous transmission '3'. Also, for the transmission of the information bits of length 3, it can be assumed that the index of the information bit set Ac2 for the position optimized on the polar code is ⁇ 2,6,7 ⁇ . In this case, the index of the information bit set Ac2 may be determined in consideration of the sequence and / or puncturing pattern used for the encoder input of the polar code.
- rate-compatibility means that the value of an existing transmitted information bit can not be changed in HARQ retransmission.
- the encoder copies the information bits located at u5 to u2, and then uses the existing u5 as an unknown frozen bit have.
- the decoder can first use the information on decoded y2.
- the encoded bits (x2, x3, x4, x5, x6, x7) A mother matrix of length 8 (i.e., Table 2) may be used for decoding.
- the code rate (R2, K / N) may be 3/8.
- the decoder may first use the information on decoded y5.
- FIG. 8 is a flowchart illustrating a method of encoding based on polar codes to support HARQ according to the present embodiment.
- the wireless device may configure a length N (N is a natural number) input vector applied to a predetermined mother matrix based on polar codes.
- the input vector is ⁇ u0, u1, ... , uN-1 ⁇ .
- the position index for the input vector is ⁇ 0,1, ... N-1 ⁇ .
- the input vector may be defined based on the first to third set of indices.
- the first set of indices may represent an optimized position of information bits of length K (K is a natural number).
- the second set of indices may represent a punctured position among coded bits generated based on the mother matrix.
- the third set of indices may indicate the location of at least one unread frozen bit generated according to an information copy technique.
- the first to third index sets are ⁇ 0, 1, ... N-1 ⁇ . ≪ / RTI >
- the order in which the output bits are decoded may be predetermined due to the structural characteristics of the mother matrix based on polar codes. For example, when N is '8', the position index is the output corresponding to ⁇ 0 ⁇ , ⁇ 4 ⁇ , ⁇ 2 ⁇ , ⁇ 6 ⁇ , ⁇ 1 ⁇ , ⁇ 5 ⁇ , ⁇ 3 ⁇
- the bits can be sequentially decoded.
- the input vector of step S810 When the input vector of step S810 is used for HARQ retransmission, the input vector may be configured to include all of the second information bits of length I (I is a natural number) for the previous transmission.
- the number of elements included in the second set of indices may be set to be reduced compared to the previous transmission.
- the first index set is ⁇ a, b, c ⁇ and the optimized position index for the second information bit for the previous transmission is ⁇ d, b, c ⁇ .
- the information contained in the position index ⁇ d ⁇ may be copied to the position index which is decoded prior to the position index ⁇ d ⁇ according to the information copy technique.
- the third index set of the input vector for HARQ retransmission may include ⁇ d ⁇ .
- the wireless device may perform encoding based on the input vector and the mother matrix.
- a first wireless device in accordance with the present embodiment may determine at least one encoded bit to be transmitted based on a second set of indexes (i.e., puncturing pattern) of N length encoded bits have. The wireless device may then transmit the at least one encoded bit to the second wireless terminal.
- a second set of indexes i.e., puncturing pattern
- the information bits can be transmitted through optimized positions while satisfying the variable code rate condition, so that the coding performance of the wireless device can be improved.
- FIG. 9 is a flowchart illustrating a procedure for performing encoding based on polar codes to support HARQ according to the present embodiment.
- K may be the length of the information bits.
- M may be the length of the codeword.
- the first wireless device acting as an encoder may determine the N and M values based on the pre-allocated resource allocation. For example, according to the above assumption, the first radio apparatus in FIG. 9 can determine N to be '8' and M to be '6'.
- the first wireless device may determine a rate-matching scheme according to the parameters for the polar code (i.e., K, N, M).
- the rate matching method may be a puncturing method or a repetition method.
- the rate matching scheme can be determined by the puncturing scheme.
- the first wireless device may determine an information bit set, a frozen bit set, a puncturing bit set, and an unannounced frozen bit set based on the parameters (i.e., K, N, M) and the rate matching scheme for polar codes have.
- the information bit set Ac2 (i.e., u2, u6, u7) for the optimized position for the information bits of size 3 in Fig. 8 may be ⁇ 2,6,7 ⁇ .
- the first wireless device of FIG. 8 may determine a frozen bit set and a puncturing bit set.
- the frozen bit set and the index of the puncturing bit set may be equal to ⁇ 0, 1 ⁇ according to the hierarchical characteristics of the mother matrix.
- the first wireless device of FIG. 8 may determine an unannounced frozen bit set.
- the index of the unannounced frozen bit set may be ⁇ 5 ⁇ .
- the unannounced frozen bits can be generated when the information bit set Ac2 at the optimized position for HARQ retransmission is different from the information bit set Ac2 at the optimized position for the previous transmission.
- the first wireless device may perform an information copy technique.
- the existing input vector u5 can be used as the unannounced frozen bits.
- the position to which the information copying technique is applied can be determined based on a predetermined decoding order based on the structural characteristics of the mother matrix.
- an input vector set i.e., (u0, u1, u2, u3, u4, u5, u6, u7)
- a mother matrix of length 8 i.e., Table 2.
- the first wireless device may determine the encoded bits to be transmitted to the second device based on the determined rate matching scheme and the puncturing bit set.
- the first wireless device may transmit the encoded bits (e.g., x2, x3, x4, x5, x6, x7 in FIG. 7) determined in step S860 to the second wireless device.
- the encoded bits e.g., x2, x3, x4, x5, x6, x7 in FIG. 7
- FIG. 10 is a flowchart illustrating a procedure for decoding based on polar codes to support HARQ according to an embodiment of the present invention.
- the second wireless device of FIG. 10 performing the role of a decoder may receive M encoded bits from the first wireless device.
- step S1020 the second wireless device of FIG. 10 may determine whether the received M coded bits are received for HARQ retransmission.
- Step S1090 will be described later.
- step S1030 If the M coded bits received by the second wireless device are not bits received for HARQ retransmission, the procedure goes to step S1030.
- the second wireless terminal of FIG. 10 may determine the K, N, and M values for the M encoded bits based on the pre-allocated resource allocation. Also, the second wireless terminal of FIG. 10 may determine the rate matching method based on the determined K, N, and M values.
- the second radio unit of FIG. 10 may be configured such that K is '3' and N is '8' based on the received encoded bits (for example, y4, y5, y6, y7 in FIG. 6) , And M can be determined as " 4 ".
- the second wireless device of FIG. 10 may determine a rate-matching scheme applied to M coded bits based on the parameters for the determined polar code (i.e., K, N, M) have.
- the second radio apparatus of FIG. 10 determines that the rate matching method is a puncturing scheme according to (K, N, M) being (3, 8, 4) .
- step S1040 the second wireless device of FIG. 10 generates an information bit set, a frozen bit set, a puncturing bit set, and an unannounced frog bit set (i.e., K, N, M) based on parameters Can be determined.
- K, N, M an unannounced frog bit set
- the information bit set Ac1 (i.e., u5, u6, u7) for the optimized position for the information bits of size 3 in Fig. 10 may be ⁇ 5,6,7 ⁇ .
- the frozen bit set and the index of the puncturing bit set may be equal to ⁇ 0, 1, 2, 3 ⁇ according to the hierarchical characteristics of the mother matrix.
- step S1050 the second wireless device of FIG. 10 may perform decoding based on the polar code.
- the second wireless apparatus of FIG. 10 can determine whether decoding is possible with polar codes smaller than N based on the hierarchical characteristics of the mother matrix.
- the second wireless apparatus of FIG. 6 A non-length 4 mother matrix (i.e., Table 1) can be used.
- step S1060 the second wireless device of FIG. 10 may determine whether the decoding operation for the received encoded bit is successful.
- the second wireless device of FIG. 10 may acquire information on the information bit of length K from the received encoded bit in step S1070.
- the second wireless device in FIG. 10 may request retransmission in step S1080.
- the second wireless device in FIG. 10 may perform LLR combining (Log-Likelihood Ratio combining).
- the encoded bits corresponding to the uncorrected frozen bits of FIG. 10 may be decoded using the bit value of the position to which the information copy technique is applied and which is first encoded.
- 11 is a conceptual diagram showing the characteristics of a permuted puncturing pattern of successive puncturing according to the present embodiment.
- a position index for an input vector (e.g., u0, ... u7) may be represented by a 3-bit binary number (b1, b2, b3).
- N may be defined as a set including modulatable permutations of length n.
- ⁇ 3 is ⁇ (1,2,3), (1,3,2), (2,1,3), (2,3,1), (3,1,2) 2, 1) ⁇ .
- ⁇ can be understood as an element of ⁇ n.
- the location index set of the puncturing pattern on which the bit-level permutation of the seed puncturing pattern of Equation (8) is performed can be defined based on Equation (9) below.
- a hierarchical puncturing pattern can be understood as a concept including both a seed puncturing pattern according to Equation (8) and a bit-level replaced puncturing pattern according to Equation (9).
- the bit-level permutation sigma 3, 2, 1
- the hierarchical puncturing pattern is expressed by Equation 8 And Equation 9 May be included.
- a short length i.e., ) Can be performed.
- FIGS. 12 and 13 are conceptual diagrams illustrating a method of encoding based on a polar code in order to support HARQ according to another embodiment of the present invention.
- the wireless device performing the role of the encoder may then transmit the encoded bits based on the index set A and the puncturing pattern ⁇ 0,0,0,1,0,1,0,1 ⁇ .
- a mother matrix of length 4 (e.g., Table 1) may be used in place of the mother matrix of 8 (e.g., Table 2).
- the index set A 'indicating the optimized position for the information bit set of FIG. 13 may be determined as ⁇ 6,7 ⁇ .
- the information bits in u5 may be copied to u6 according to the information copying technique.
- u6 is a position where u6 is decoded before u5 in the decoding process.
- u5 can be used as unannounced frozen bits.
- the information bits can be transmitted through the index set A 'while satisfying the variable code rate condition.
- the m puncturing patterns can be defined as p (1), p (2), ..., p (m).
- the puncturing pattern can be defined as the following Equation (10).
- the information set A can be designed using a punctured polar sign having the highest code rate (i.e., a code rate rm)
- a non-zero frozen vector can be designed by applying the optimized information sets and the algorithm shown in Table 4 below.
- the encoding process for the variable-code-rate polynomial code according to the present embodiment may be the same as the existing one.
- a non-zero frozen vector (or an information-bit dependent frozen vector) obtained from Table 4 below can be used.
- the information bits can be copied to a portion of the frozen vector by the information-copy method.
- the frozen vector according to the present embodiment is not always an all-zero vector but can be determined depending on the information-bit.
- the algorithm of Table 3 can be performed only once, but the algorithm of Table 4 below can be performed every initial transmission in HARQ-IR to obtain a non-zero frozen vector (or information-dependent frozen vector).
- FIG. 14 shows a simulation result of a polar code designing a variable code rate according to the present embodiment.
- the length K of the information bits is '52', and K includes 8 bits of CRC bits.
- the coding rate can be assumed to be r1 (52/256) ⁇ r2 (52/192) ⁇ r3 (52/192) ⁇ r4 (52/64).
- the puncturing pattern can be determined according to Equation (11) below.
- the punctured poles with the highest code rate, r4 are transmitted. Then, in the second retransmission, a redundancy coded bit corresponding to the indexes belonging to Bp (4) -Bp (3) may be transmitted.
- the signals received in the first and second transmissions may then be combined to form a punctured polar code with a code rate of r3.
- redundant coded bits corresponding to Bp (3) -Bp (2) may be transmitted.
- redundant coded bits corresponding to Bp (2) -Bp (1) may be transmitted. That is, if all the received signals up to the fourth transmission are combined, a mother pole code with a code rate r1 can be constructed.
- FER Frame-Error-Rate
- 15 is a diagram for explaining wireless devices to which the present exemplary embodiments can be applied.
- the base station apparatus 10 of FIG. 15 may include a receiving module 11, a transmitting module 12, a processor 13, a memory 14, and a plurality of antennas 15.
- the receiving module 11 can receive various signals, data and information from an external device (e.g., a terminal).
- the transmitting module 12 can transmit various signals, data, and information to an external device (e.g., a terminal).
- the receiving module 11 and the transmitting module 12 may be included in one transceiver.
- the processor 13 may control the operation of the overall base station apparatus 10.
- the processor 13 may also perform a function of calculating information received by the base station apparatus 10 and information to be transmitted to the outside by the base station apparatus 10.
- the memory 14 may store the arithmetic processed information for a predetermined time.
- the plurality of antennas 15 are configured to support MIMO transmission / reception, and may be configured according to a two-dimensional antenna arrangement.
- the terminal device 20 of FIG. 15 may include a receiving module 21, a transmitting module 22, a processor 23, a memory 24, and a plurality of antennas 25.
- the receiving module 11 can receive various signals, data and information from an external device (e.g., a terminal).
- the transmitting module 12 can transmit various signals, data, and information to an external device (e.g., a terminal).
- the receiving module 11 and the transmitting module 12 may be included in one transceiver.
- the processor 13 may control the operation of the overall base station apparatus 10.
- the processor 13 may also perform a function of calculating information received by the base station apparatus 10 and information to be transmitted to the outside by the base station apparatus 10.
- the memory 14 may store the arithmetic processed information for a predetermined time.
- the plurality of antennas 15 are configured to support MIMO transmission / reception, and may be configured according to a two-dimensional antenna arrangement.
- the downlink transmission entity or the uplink receiving entity is mainly described as a base station, and the downlink receiving entity or the uplink transmitting entity is mainly described as a terminal.
- the scope of the present specification is not limited thereto.
- the above-described embodiments can be implemented by various means.
- the above-described embodiments may be implemented by hardware, firmware, software, or a combination thereof.
- the embodiments described above may be implemented in one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs) Programmable Gate Arrays, a processor, a controller, a microcontroller, a microprocessor, and the like.
- ASICs application specific integrated circuits
- DSPs digital signal processors
- DSPDs digital signal processing devices
- PLDs programmable logic devices Programmable Gate Arrays
- processor a controller, a microcontroller, a microprocessor, and the like.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Probability & Statistics with Applications (AREA)
- Theoretical Computer Science (AREA)
- Detection And Prevention Of Errors In Transmission (AREA)
- Error Detection And Correction (AREA)
Abstract
본 실시 예에 따른 HARQ를 지원하기 위해 폴라 코드를 기반으로 부호화하는 방법은, 무선 장치가, 폴라 코드를 기반으로 미리 결정된 마더 행렬에 적용되는 재전송을 위한 길이 N의 입력 벡터를 구성하되, 입력 벡터는 길이 K의 정보 비트의 최적화된 위치를 위한 제1 인덱스 집합, 마더 행렬을 기반으로 생성되는 부호화된 비트 중 천공될 위치에 대한 제2 인덱스 집합 및 정보 복사 기법과 연관된 적어도 하나의 언노운 프로즌 비트의 위치에 대한 제3 인덱스 집합을 기반으로 정의되는, 단계; 및 무선 장치가, 입력 벡터 및 마더 행렬을 기반으로 부호화를 수행하는 단계를 포함한다.
Description
본 명세서는 무선 통신에 관한 것으로, 더 상세하게는 HARQ를 지원하기 위해 폴라 코드를 기반으로 부호화하는 방법 및 이를 이용한 무선 장치에 관한 것이다.
데이터 통신 시스템에서 송신자로부터 수신자까지 데이터를 오류 없이 전달하는 것은 중요한 문제이다. 1948년 쉐논(Shannon)이 오류 없이 전달 가능한 최대 데이터 전송률의 한계를 수학적으로 규명 했는데, 이를 채널 용량 (channel capacity) 라고 한다. 이러한 채널 용량에 가깝게 실제 통신 시스템을 구현하기 위해서는, 구현 가능한 복잡도를 가지는 오류 정정 부호 (error correction code)가 존재하여야 한다. 1948년 이후 여러 종류의 오류 정정 부호가 개발되었으며, 비교적 최근 쉐논의 채널 용량에 근접한 성능을 보이는 오류 정정 부호로 터보 코드(turbo code)와 LDPC(Low Density Parity Check) 등이 개발되었다. 그러나, 이와 같은 코드들은 쉐논의 채널 용량에 근접한 성능을 보이기는 하나, 정확한 채널 용량을 달성하는 것이 아니다. 최근 이와 같은 문제점을 해결하면서 수학적으로 채널 용량을 완전히 만족하는 코드로 폴라 코드(polar code)가 개발되었다.
HARQ는 오류가 있는 패킷을 수신하면 재전송을 요청함으로써 오류를 복원하는 기술이다.
본 명세서의 목적은 HARQ를 지원하기 위해 폴라 코드를 기반으로 부호화하는 방법 및 이를 이용한 무선 장치를 제공하는데 있다.
본 실시 예에 따른 HARQ를 지원하기 위해 폴라 코드를 기반으로 부호화하는 방법은, 무선 장치가, 폴라 코드를 기반으로 미리 결정된 마더 행렬에 적용되는 재전송을 위한 길이 N의 입력 벡터를 구성하되, 입력 벡터는 길이 K의 정보 비트의 최적화된 위치를 위한 제1 인덱스 집합, 마더 행렬을 기반으로 생성되는 부호화된 비트 중 천공될 위치에 대한 제2 인덱스 집합 및 정보 복사 기법과 연관된 적어도 하나의 언노운 프로즌 비트의 위치에 대한 제3 인덱스 집합을 기반으로 정의되는, 단계; 및 무선 장치가, 입력 벡터 및 마더 행렬을 기반으로 부호화를 수행하는 단계를 포함한다.
본 명세서의 목적은 향상된 성능으로 HARQ를 지원하기 위해 폴라 코드를 기반으로 부호화하는 방법 및 이를 이용한 무선 장치를 제공하는데 있다.
도 1은 폴라 부호화(polar encoding)의 개념을 도시하는 도면이다.
도 2는 코드 블록의 길이 에 따른 양극화 경향을 도시하는 도면이다.
도 3은 폴라 코드의 기초 모듈(base module)을 도시하는 도면이다.
도 4는 N 레벨을 갖는 채널 조합의 예시에 관한 도면이다.
도 5는 본 일 실시 예에 따른 천공에 따른 상호(reciprocal) 특성에 관한 개념도를 보여준다.
도 6은 본 일 실시 예에 따른 천공에 따른 계층(hierarchical) 특성에 기반한 천공 패턴에 관한 개념도이다.
도 7은 본 일 실시 예에 따른 HARQ를 지원하기 위해 폴라 코드를 기반으로 부호화하는 방법에 관한 개념도이다.
도 8은 본 일 실시 예에 따른 HARQ를 지원하기 위해 폴라 코드를 기반으로 부호화하는 방법에 관한 순서도이다.
도 9는 본 일 실시 예에 따라 HARQ를 지원하기 위해 폴라 코드를 기반으로 부호화를 수행하는 절차에 관한 순서도이다.
도 10는 본 일 실시 예에 따라 HARQ를 지원하기 위해 폴라 코드를 기반으로 복호화를 수행하는 절차에 관한 순서도이다.
도 11은 본 일 실시 예에 따른 연속 펑처링(successive puncturing)의 치환된 천공 패턴(permuted punctiuring pattern)의 특성을 보여주는 개념도이다.
도 12 및 도 13은 본 다른 실시 예에 따라 HARQ를 지원하기 위해 폴라 코드를 기반으로 부호화하는 방법에 대한 개념도이다.
도 14는 본 일 실시 예에 따른 가변 부호율을 지원하는 극 부호 설계에 대한 시뮬레이션 결과를 보여준다.
도 15는 본 일 실 시 예들이 적용될 수 있는 무선 장치들을 설명하기 위한 도면이다.
전술한 특성 및 이하 상세한 설명은 모두 본 명세서의 설명 및 이해를 돕기 위한 예시적인 사항이다. 즉, 본 명세서는 이와 같은 실시 예에 한정되지 않고 다른 형태로 구체화될 수 있다. 다음 실시 형태들은 단지 본 명세서를 완전히 개시하기 위한 예시이며, 본 명세서가 속하는 기술 분야의 통상의 기술자들에게 본 명세서를 전달하기 위한 설명이다. 따라서, 본 명세서의 구성 요소들을 구현하기 위한 방법이 여럿 있는 경우에는, 이들 방법 중 특정한 것 또는 이와 동일성 있는 것 가운데 어떠한 것으로든 본 명세서의 구현이 가능함을 분명히 할 필요가 있다.
본 명세서에서 어떤 구성이 특정 요소들을 포함한다는 언급이 있는 경우, 또는 어떤 과정이 특정 단계들을 포함한다는 언급이 있는 경우는, 그 외 다른 요소 또는 다른 단계들이 더 포함될 수 있음을 의미한다. 즉, 본 명세서에서 사용되는 용어들은 특정 실시 형태를 설명하기 위한 것일 뿐이고, 본 명세서의 개념을 한정하기 위한 것이 아니다. 나아가, 발명의 이해를 돕기 위해 설명한 예시들은 그것의 상보적인 실시 예도 포함한다.
본 명세서에서 사용되는 용어들은 본 명세서가 속하는 기술 분야의 통상의 기술자들이 일반으로 이해하는 의미를 갖는다. 보편적으로 사용되는 용어들은 본 명세서의 맥락에 따라 일관적인 의미로 해석되어야 한다. 또한, 본 명세서에서 사용되는 용어들은, 그 의미가 명확히 정의된 경우가 아니라면, 지나치게 이상적이거나 형식적인 의미로 해석되지 않아야 한다. 이하 첨부된 도면을 통하여 본 명세서의 실시 예가 설명된다.
도 1은 폴라 부호화(polar encoding)의 개념을 도시하는 도면이다.
도 1을 참조하면, 폴라 코드를 이용한 폴라 부호화 과정은 채널 조 합(channel combining) 단계(S1) 및 채널 분할(channel splitting) 단계(S2)를 포함할 수 있다.
도 1을 참조하면, 복수의 기본 채널(W)을 적절히 조합하여 백터 채널(Wvcc)이 생성 될 수 있다. 또한, 생성된 백터 채널(Wvcc)을 분리함으로써, 양극화된(polarlized) 새로운 채널들이 생성될 수 있다. 예를 들어, 무한한 길이의 채널에 대하여 채널 용량은 채널 용량 C(W)=0과 C(W)=1로 구분될 수 있다.
예를 들어, 채널 용량(C(W))이 '1'인 채널은 오류 없는 전송이 가능하므로 정보 비트(information bit)의 전송을 위해 할당될 수 있다. 채널 용량(C(W))이 '0'인 채널은 정보 전송이 불가능하므로 의미 없는 비트인 프로즌 비트(frozen bit)의 전송을 위해 할당될 수 있다.
도 2는 코드 블록의 길이 에 따른 양극화 경향을 도시하는 도면이다.
도 2의 (a) 내지 (c)를 참조하면, 코드 블록의 길이(N)가 길어질수록 채널 용량이 1 또는 0으로 양극화됨은 이해될 것이다.
도 3은 폴라 코드의 기초 모듈(base module)을 도시하는 도면이다.
도 1 내지 도 3을 참조하면, 상술한 채널 조합은 B-DMC 채널(binary-input discrete memoryless channel)을 병렬(parallel)로 연접(permutation)함으로써 수행될 수 있다.
예를 들어, 채널 조합을 기반으로 코드 블록의 크기가 결정 될 수 있다. 예를 들어, 도 3의 W는 B-DMC 채널을 나타낼 수 있다. 도 3의 폴라 코드의 기초 모듈은 2개의 B-DMC 채널들을 조합할 수 있다.
예를 들어, 도 3의 u1 및 u2는 이진-입력 소스 비트(binary-input source bit)일 수 있다. 도 3의 x1 및 x2는 상응하는 채널(W)로의 입력일 수 있다. 도 3의 y1 및 y2는 부호화된 출력 비트(output coded bit)일 수 있다. 또한, 전체 채널과 동등한(equivalent) 채널은 W2로 표시될 수 있다.
예를 들어, x1에는 u1 및 u2의 XOR 연산의 결과 값이 포함될 수 있다. 또한, x2에는 u2가 별도의 연산 없이 포함될 수 있다. 결국, 도 3의 폴라 코드의 기초 모듈에 대한 기초 행렬(F)는 하기의 수학식 1과 같이 표현될 수 있다.
만일 N개의 B-DMC 채널이 조합될 때, 각각의 채널은 재귀적(recursive) 형태로 표현될 수 있다. 또한, 인코더의 출력 비트에 대한 x 벡터는 하기의 수학식 2를 이용하여 표현될 수 있다.
위 수학식 2를 참조하면, 인코더의 출력 비트에 대한 x 벡터는 하기의 수학식 3과 같이 표현되고, 인코더의 입력 비트에 대한 u 벡터는 하기의 수학식 4와 같이 표현될 수 있다.
예를 들어, 코드 블록의 길이가 N인 폴라 코드에 있어서, K 비트의 입력을 가정하면, 부호율(code rate)은 N/K이다. 이 때, {u1, …uN} 중에서 K개의 비트는 페이로드를 위한 비트로 이해될 수 있다.
또한, 위 수학식 2를 참조하면, 하기 수학식 5의 생성 행렬(GN)은 수학식 1의 기초 행렬(F)의 크로네커 곱의 형태로 표현될 수 있다.
예를 들어, 수학식 5 및 수학식 6에 따라 2X2 크기의 생성 행렬(G1)은 수학식 1에 상응할 수 있다.
또한, 수학식 5 및 수학식 6에 따라4X4 크기의 생성 행렬(G2)은 하기의 표 1과 같을 수 있다.
또한, 수학식 5 및 수학식 6에 따라8X8 크기의 생성 행렬(G3)은 하기의 표 2과 같을 수 있다.
본 명세서에서, 위 수학식 5에 따른 생성행렬(GN)은 폴라 코드를 기반으로 미리 결정된 마더 행렬(mother matrix)로 언급될 수 있다.
또한, 본 명세서에서, 하기 설명될 도 4의 경우와 달리, Bit-reverse permutation을 위한 연산이 인코딩 파트가 아닌 디코딩 파트에 존재하는 것을 전제로 설명된다.
도 4는 N 레벨을 갖는 채널 조합의 예시에 관한 도면이다. 예를 들어, 도 4는 Bit-reverse permutation을 위한 연산이 인코딩 파트에 존재하는 경우로 이해될 수 있다. 도 4를 참조하면, 코드 블록의 크기(N)은 2^n(n은 자연수)의 제한을 가질 수 있다.
예를 들어, 폴라 코드의 복호화는 연속 제거(Successive Cancellation, SC) 복호화를 이용하여 수행될 수 있다. 여기서, 연속 제거 복호화 방식은 채널 천 이(transition) 확률을 계산한 후, 계산된 확률에 기초하여 입력 비트에 대한 우도비 (likelihood Ratio, LLR)를 계산함으로써 수행될 수 있다.
이 경우, 채널 조합과 채널 분할 과정이 재귀적 형태인 점을 이용하여, 채널 천이 확률은 재귀적인 형태로 계산될 수 있다. 즉, 우도비 값도 재귀적인 형태로 계산될 수 있다.
이하, 데이터가 전송되는 페이로드 비트는 데이터 비트로, 페이로드 비트를 제외한 나머지 비트(즉, 실제 데이터가 전송되지 않는 비트)는 프로즌 비트(frozen bit)로 언급될 수 있다.
폴라 코드에 있어서, 코드 블록 내 에서 어떠한 위치의 비트를 데이터 비트 또는 프로즌 비트로 결정하는지가 문제될 수 있다. 예를 들어, 각 채널의 채널 용량에 기초하여 데이터 비트의 위치와 프로즌 비트의 위치가 결정될 수 있다.
도 5는 본 일 실시 예에 따른 천공에 따른 상호(reciprocal) 특성에 관한 개념도를 보여준다.
본 명세서에서, 생성 행렬(GN) 즉, 마더 행렬에 적용될 입력 벡터는 {u0,u1, …,um-1, m은 자연수}로 정의될 수 있다.
예를 들어, 표 2의 8X8 크기의 생성 행렬(G3)를 위한 입력 벡터는 {u0,u1, …,u7}로 정의될 수 있다. 또한, 입력 벡터를 위한 위치 인덱스는 {0,1, …N-1}로 정의될 수 있다.
도 5를 참조하면, 길이 8의 입력 벡터(예로, u0, … u7)는 수학식 2에 따라 길이 8의 폴라 코드를 기반으로 부호화된 벡터 집합(예로, x0, …x7)으로 부호화될 수 있다.
예를 들어, 부호화된 벡터 집합{x0, …x7}이 각 채널(W)을 통과하면, 수신단에서 {y0, …y7}이 수신될 수 있음은 이해될 것이다.
본 명세서에서, 천공 패턴(puncturing pattern, PN)은 하기의 수학식 7과 같이 정의될 수 있다.
여기서, Pi=0은 i번째 부호화된 비트(coded bit)가 전송되지 않음을 의미할 수 있다. 예를 들어, 도 5과 같이 천공 패턴이 PN=(0,0,0,0,1,1,1,1)로 표현될 때, 천공된 부호화된 비트들(punctured coded bits)의 위치 인덱스는 B_PN={0,1,2,3}로 표현될 수 있다.
즉, B_PN={0,1,2,3}와 연관된 입력 벡터(즉, u0,u1,u2,u3)에 상응하는 채널을 통해 어떠한 정보도 전달될 수 없다. 다시 말해, B_PN={0,1,2,3}와 연관된 입력 벡터(즉, u0,u1,u2,u3) 각각은 제로 용량(zero capacity)를 갖는 채널로 이해될 수 있다.
본 명세서에서 언급되는 천공에 따른 상호 특성을 기반으로 도 5와 같이 제로 용량(zero capacity)을 갖는 채널들(예로, u0 내지 u3)의 인덱스와 천공된 비트들(punctured bits, 예로, y0,y1,y2y3)의 인덱스는 동일할 수 있다.
나아가, 천공에 따른 상호 특성에 따라 천공된 비트들(punctured bits, 예로, y0,y1,y2y3)에 상응하는 입력 벡터(예로, u0,u1,u2,u3)은 알려진(known) 값을 갖는 프로즌 벡터 집합으로 설정될 수 있다.
도 6은 본 일 실시 예에 따른 천공에 따른 계층(hierarchical) 특성에 기반한 천공 패턴에 관한 개념도이다.
도 5 및 도 6을 참조하면, 위치 인덱스 {0}부터 {3}까지 연속적으로(successive) 천공될 때, 연속적인 천공 패턴은 계층 특성이 반영된 천공 패턴으로 이해될 수 있다.
도 6의 명확하고 간결한 이해를 위해, B_PN={0,1,2,3}에 상응하는 입력 벡터(예로, u0,u1,u2,u3) 각각은 제로 용량(zero capacity)를 갖는 채널이라고 가정할 수 있다.
도 6의 폴라 코드를 위한 마더 행렬의 크기는 '8'이고, B_PN={0,1,2,3}에 따른 코드워드의 크기는 '4'일 수 있다. 이 경우, 정보 비트의 크기는 '3'으로 가정할 수 있다.
또한, 길이 3의 정보 비트의 전송을 위하여, 폴라 코드 상에서 최적화된 위치에 관한 정보 비트 집합(Ac1)의 인덱스는 {5,6,7}라고 가정할 수 있다. 예를 들어, 정보 비트 집합(Ac1)의 인덱스는 폴라 코드의 인코더 입력에 이용되는 시퀀스 및/또는 천공 패턴을 고려하여 결정될 수 있다.
예를 들어, 언노운(unknown) 비트인 'a'가 u5에 위치되고, 언노운 비트인 'b'가 u6에 위치되고, 언노운 비트인 'c'가 u7에 위치될 수 있다.
본 명세서에서 시드 천공 패턴(seed puncturing pattern)이 언급될 수 있다. 예를 들어, 시드 천공 패턴의 비트 인덱스 집합은 하기의 수학식 8을 통해 정의될 수 있다.
도 6을 참조하면, u0에 상응하는 x0가 천공될 때, u1 내지 u7에 상응하는 부호화된 비트 값들에 영향이 없음은 이해될 것이다.
다시 말해, 본 일 실시 예에 따른 계층 특성에 따라, 상대적으로 위에 위치된 입력 값의 변화는 상대적으로 아래에 위치된 입력 값에 대한 출력 값에 영향을 줄 수 없다.
도 6의 실시 예에 따른 인코더는 천공 패턴(예로, PN=(0,0,0,0,1,1,1,1))을 고려하여, 부호화된 비트 x0 내지 x7 중 천공 패턴이 '0'인 위치의 값은 전송하지 않을 수 있다.
도 6을 참조하면, 전송되는 부호화된 비트들(예로, x4,x5,x6,x7)은 천공 전의 폴라 코드 사이즈보다 작은 사이즈를 갖는 폴라 코드를 이용하여 복호화될 수 있다.
예를 들어, 도 6의 전송되는 부호화된 비트들(예로, x4,x5,x6,x7)은 디코딩 단계에서 길이 8의 마더 행렬(즉, 표 2)을 대신하여 길이 4의 마더 행렬(즉, 표 1)이 이용될 수 있다. 이 경우, 수신 관점에서, 부호율(R1, K/N)은 3/4일 수 있다.
도 7은 본 일 실시 예에 따른 HARQ를 지원하기 위해 폴라 코드를 기반으로 부호화하는 방법에 관한 개념도이다.
도 5 내지 도 7을 참조하면, 표 2의 8X8 크기의 마더 행렬(G3)을 위한 입력 벡터는 {u0,u1, …,u7}로 정의될 수 있다. 즉, 도 7의 입력 벡터를 위한 위치 인덱스는 {0,1, …}로 정의될 수 있다.
예를 들어, 도 7의 천공 패턴의 위치 인덱스는 B_PN={0,1}이고, 계층 특성이 반영된 천공 패턴으로 이해될 수 있다.
본 일 실시 예에 따르면, 출력 비트 집합(예, y0, …y7)의 복호되는 순서는 미리 정해질 수 있다. 예를 들어, u0, u4, u2, u6, u1, u5, u3, u7의 순서에 따라 부호화된 출력 값(즉, y0, y4, y2, y6, y1, y5, y3, y7)은 순차적으로 복호될 수 있다.
도 7의 명확하고 간결한 이해를 위해, 도 7은 도 6에 해당하는 이전의 전송의 실패한 후 HARQ 재전송인 케이스를 전제로 설명된다. 즉, 도 7의 입력 벡터 {u0,u1, …,u7}는 HARQ 재전송을 위해 구성될 수 있다.
예를 들어, B_PN={0,1}에 상응하는 입력 벡터(예로, u0,u1) 각각은 제로 용량(zero capacity)를 갖는 채널로 이해될 수 있다. 예를 들어, 도 7의 폴라 코드를 위한 마더 행렬의 크기는 도 6과 동일한 '8'이고, 도 7의 코드워드 크기는 HARQ 재전송을 고려하여 이전의 전송보다 확장된 '6'일 수 있다.
또한, 정보 비트의 크기는 이전의 전송과 동일한 '3'으로 가정할 수 있다. 또한, 길이 3의 정보 비트의 전송을 위하여, 폴라 코드 상에서 최적화된 위치에 관한 정보 비트 집합(Ac2)의 인덱스는 {2,6,7}이라고 가정할 수 있다. 이 경우, 정보 비트 집합(Ac2)의 인덱스는 폴라 코드의 인코더 입력에 이용되는 시퀀스 및/또는 천공 패턴을 고려하여 결정될 수 있다.
본 명세서에서, 가변 부호율 조건(rate-compatibility)은 기존 전송된 정보 비트의 값이 HARQ 재전송에서 변화될 수 없음을 의미한다.
예를 들어, 도 7에 따른 HARQ 재전송을 위해 이전의 전송에서 이용된 도 6의 Ac1={5,6,7}에 상응하는 입력 벡터(예로, u5,u6,u7)에는 반드시 정보 비트가 포함되어야 한다.
그러나, 위 가정에 따라, 도 7에서 최적화된 정보 비트 집합(Ac2)의 인덱스는 {2,6,7}이므로, 가변 부호율 조건만을 고려하여 전송을 수행하면, 성능이 저하될 수 있다.
도 7의 일 실시 예에 따르면, 정보 복사(information copy) 기법이 적용됨에 따라, 인코더는 u5에 위치한 정보 비트를 u2로 카피한 후, 기존의 u5을 언노운 프로즌 비트(unknown frozen bit)로 사용할 수 있다.
이 경우, 디코딩 관점에서, y5가 복호될 때, 디코더는 먼저 복호화된 y2에 대한 정보를 이용할 수 있다.
또한, 도 7의 전송되는 부호화된 비트들(x2,x3,x4,x5,x6,x7)의 길이(6)를 고려할 때, 부호화된 비트들(x2,x3,x4,x5,x6,x7)의 복호화를 위해서 길이 8의 마더 행렬(즉, 표 2)이 이용될 수 있다. 이 경우, 수신 관점에서, 부호율(R2, K/N)은 3/8일 수 있다.
도 7을 참조하면, 정보 복사 기법이 HARQ 재전송에서 적용될 때, 가변 부호율 조건을 만족하면서 Ac2={2,6,7}에 상응하는 위치에 정보 비트가 포함될 수 있다.
다만, 본 명세서가 도 7의 실시 예에 한정되는 것이 아님은 이해될 것이다. 예를 들어, N=8인 경우, 일반적으로 u2보다 u3의 채널 용량(channel capacity)이 높기 때문에, u2와 u3 중 u2부터 frozen 비트로 사용된다. 이러한 특성을 고려할 때, u5에 위치한 정보 비트가 u3로 복사될 수도 있다.
이 경우, 디코딩 관점에서, y3가 복호될 때, 디코더는 먼저 복호화된 y5에 대한 정보를 이용할 수도 있다.
도 8은 본 일 실시 예에 따른 HARQ를 지원하기 위해 폴라 코드를 기반으로 부호화하는 방법에 관한 순서도이다.
S810 단계에서, 무선 장치는 폴라 코드를 기반으로 미리 결정된 마더 행렬(mother matrix)에 적용되는 길이 N(N은 자연수)의 입력 벡터를 구성할 수 있다. 입력 벡터는 {u0,u1, …,uN-1}로 정의될 수 있다. 예를 들어, 입력 벡터를 위한 위치 인덱스는 {0,1, …N-1}일 수 있다.
여기서, 입력 벡터는 제1 내지 제3 인덱스 집합을 기반으로 정의될 수 있다. 예를 들어, 제1 인덱스 집합은 길이 K(K는 자연수)의 정보 비트의 최적화된 위치를 나타낼 수 있다.
예를 들어, 제2 인덱스 집합은 마더 행렬을 기반으로 생성되는 부호화된 비트 중 천공될 위치를 나타낼 수 있다.
예를 들어, 제3 인덱스 집합은 정보 복사(information copy) 기법에 따라 생성되는 적어도 하나의 언노운 프로즌 비트의 위치를 나타낼 수 있다.
다시 말해, 제1 인덱스 집합 내지 제3 인덱스 집합은 {0,1, …N-1}을 기반으로 정의될 수 있다.
본 일 실시 예에 따르면, 출력 비트의 복호화되는 순서는 폴라 코드에 기반한 마더 행렬의 구조적 특징으로 인하여 미리 결정될 수 있다. 예를 들어, N이 '8'일 때, 위치 인덱스는 {0}, {4}, {2}, {6}, {1}, {5}, {3}, {7}에 상응하는 출력 비트들이 순차적으로 복호될 수 있다.
S810 단계의 입력 벡터가 HARQ 재전송을 위해 이용될 때, 입력 벡터는 이전의 전송을 위한 길이 I(I는 자연수)의 제2 정보 비트를 모두 포함하도록 구성될 수 있다.
또한, S810 단계의 입력 벡터가 HARQ 재전송을 위해 이용될 때, 제2 인덱스 집합에 포함된 원소들의 개수는 이전의 전송보다 감소되도록 설정될 수 있다.
예를 들어, K 및 I가 모두 '3'일 때, 제1 인덱스 집합은 {a,b,c}이고, 이전의 전송을 위한 제2 정보 비트를 위한 최적화된 위치 인덱스는 {d,b,c}이라고 가정할 수 있다.
즉, 위치 인덱스 {d}에 포함된 정보는 정보 복사 기법에 따라 위치 인덱스 {d}보다 먼저 디코딩되는 위치 인덱스로 복사될 수 있다.
이 경우, 위치 인덱스 {d}가 위치 인덱스 {a}로 복사될 때, HARQ 재전송을 위한 입력 벡터의 제3 인덱스 집합은 {d}를 포함할 수 있다.
S820 단계에서, 무선 장치는 입력 벡터 및 마더 행렬을 기반으로 부호화를 수행할 수 있다.
본 도면에 도시되지 않으나, 본 일 실시 예에 따른 제1 무선 장치는 N 길이의 부호화된 비트 중 제2 인덱스 집합(즉, 천공 패턴)을 기반으로 전송될 적어도 하나의 부호화된 비트를 판단할 수 있다. 이어, 무선 장치는 적어도 하나의 부호화된 비트를 제2 무선 단말로 전송할 수 있다.
결국, 본 명세서에 따르면, HARQ 재전송을 수행할 때, 가변 부호율 조건을 만족하면서 동시에 최적화된 위치를 통해 정보 비트를 전송할 수 있으므로, 무선 장치의 부호화 성능이 향상될 수 있다.
도 9는 본 일 실시 예에 따라 HARQ를 지원하기 위해 폴라 코드를 기반으로 부호화를 수행하는 절차에 관한 순서도이다.
이하, 언급되는 폴라 코드를 위한 파라미터(즉, K, N, M)는 다음과 같이 정의될 수 있다. 예를 들어, K는 정보 비트의 길이일 수 있다. N(=2^n, n은 자연수)은 폴라코드를 위한 마더 행렬(이하, 마더 폴라 코드)의 크기일 수 있다. 또한, M은 코드워드의 길이일 수 있다.
도 9의 명확하고 간결한 설명을 위하여, 폴라 코드를 위한 파라미터(즉, K, N, M)는 도 7과 같이 (3, 8, 6)으로 가정할 수 있다.
S910 단계에서, 정보 비트의 길이(K)가 '3'인 경우, 인코더의 역할을 수행하는 제1 무선 장치는 미리 할당된 자원 할당을 기반으로 N 및 M 값을 결정할 수 있다. 예를 들어, 위 가정에 따라, 도 9의 제1 무선 장치는 N은 '8'이고, M은 '6'로 결정할 수 있다.
S920 단계에서, 제1 무선 장치는 폴라 코드를 위한 파라미터(즉, K, N, M)에 따라 레이트 매칭(rate-matching) 방식을 결정할 수 있다. 이 경우, 레이트 매칭 방식은 펑처링(puncturing) 방식 또는 반복(repetition) 방식이 있을 수 있다.
예를 들어, 폴라 코드를 위한 파라미터(즉, K, N, M)는 (3, 8, 6)이므로, 레이트 매칭 방식은 펑처링 방식으로 결정될 수 있다.
S930 단계에서, 제1 무선 장치는 폴라 코드를 위한 파라미터(즉, K, N, M) 및 레이트 매칭 방식을 기반으로 정보 비트 집합, 프로즌 비트 집합, 펑처링 비트 집합 및 언노운 프로즌 비트 집합을 결정할 수 있다.
예를 들어, 도 8의 크기 3의 정보 비트를 위한 최적화된 위치에 관한 정보 비트 집합(Ac2, 즉, u2, u6, u7)은 {2,6,7}일 수 있다.
예를 들어, 도 8의 제1 무선 장치는 프로즌 비트 집합 및 펑처링 비트 집합을 결정할 수 있다. 일 예로, 도 7과 같이, 마더 행렬의 계층 특성에 따라 프로즌 비트 집합 및 펑처링 비트 집합의 인덱스는 {0,1}로 동일할 수 있다.
예를 들어, 도 8의 제1 무선 장치는 언노운 프로즌 비트 집합을 결정할 수 있다. 일 예로, 도 7과 같이, 언노운 프로즌 비트 집합의 인덱스는 {5}일 수 있다.
다시 말해, 언노운 프로즌 비트는 HARQ 재전송을 위한 최적화된 위치에 정보 비트 집합(Ac2)이 이전의 전송을 위한 최적화된 위치의 정보 비트 집합(Ac2)이 다른 경우에 생성될 수 있다.
S940 단계에서, 제1 무선 장치는 정보 복사(information copy) 기법을 수행할 수 있다.
예를 들어, 도 7과 같이, 입력 벡터인 u5에 위치한 정보 비트는 입력 벡터인 u2로 카피된 후, 기존 입력 벡터인 u5는 언노운 프로즌 비트로 사용될 수 있다.
전술한 바와 같이, 정보 복사 기법이 적용되는 위치는 마더 행렬의 구조적 특징을 기반으로 미리 정해진 디코딩 순서를 기반으로 결정될 수 있다.
S950 단계에서, 제1 무선 장치는 폴라 코드를 기반으로 부호화를 수행할 수 있다. 예를 들어, 도 7을 참조하면, 제1 무선 장치는 입력 벡터 집합(즉, (u0,u1,u2,u3,u4,u5,u6,u7)=(0,0,u2,u5,u4,u5,u6,u7)) 및 길이 8의 마더 행렬(즉, 표 2)을 이용하여 부호화를 수행할 수 있다.
S960 단계에서, 제1 무선 장치는 결정된 레이트 매칭 방식 및 펑처링 비트 집합을 기반으로 제2 장치로 전송될 부호화된 비트를 결정할 수 있다.
예를 들어, 도 7과 같이 길이 8의 부호화된 비트들(x0, …x7) 중에서 천공 패턴(즉, PN=(0,0,1,1,1,1,1,1))에 따라 길이 6의 부호화된 비트들(x2,x3,x4,x5,x6,x7)은 전송될 부호화된 비트로 결정될 수 있다.
S970 단계에서, 제1 무선 장치는 S860 단계에서 결정된 부호화된 비트(예로, 도 7의 x2,x3,x4,x5,x6,x7)를 제2 무선 장치로 전송할 수 있다.
도 10는 본 일 실시 예에 따라 HARQ를 지원하기 위해 폴라 코드를 기반으로 복호화를 수행하는 절차에 관한 순서도이다.
도 9 및 도 10을 참조하면, S1010 단계에서, 디코더의 역할을 수행하는 도 10의 제2 무선 장치는 M개의 부호화된 비트들을 제1 무선 장치로부터 수신할 수 있다.
S1020 단계에서, 도 10의 제2 무선 장치는 수신된 M개의 부호화된 비트들이 HARQ 재전송을 위해 수신된 것인지 여부를 판단할 수 있다.
만일 제2 무선 장치에 수신된 M개의 부호화된 비트들이 HARQ 재전송을 위해 수신된 비트라고 판단되면, 수순은 S1090 단계로 진행된다. S1090 단계는 후술된다.
만일 제2 무선 장치에 수신된 M개의 부호화된 비트들이 HARQ 재전송을 위해 수신된 비트가 아닌 경우, 수순은 S1030 단계로 진행된다.
S1030 단계에서, 도 10의 제2 무선 단말은 미리 할당된 자원 할당을 기반으로 M개의 부호화된 비트들에 대한 K, N 및 M 값을 판단할 수 있다. 또한, 도 10의 제2 무선 단말은 판단된 K, N 및 M 값을 기반으로 레이트 매칭 방식을 판단할 수 있다.
예를 들어, 도 10의 제2 무선 장치는 도 6과 같이 수신된 부호화된 비트(예로, 도 6의 y4,y5,y6,y7)를 기반으로 K는 '3'이고, N은 '8'이고, M은 '4'로 판단할 수 있다.
예를 들어, 도 10의 제2 무선 장치는 결정된 폴라 코드를 위한 파라미터(즉, K, N, M)을 기반으로 M개의 부호화된 비트들에 적용된 레이트 매칭(rate-matching) 방식을 판단할 수 있다.
예를 들어, 도 10의 제2 무선 장치는 도 6과 같이 (K, N, M)가 (3, 8, 4)임에 따라, 제2 무선 장치는 레이트 매칭 방식을 펑처링 방식으로 판단할 수 있다.
S1040 단계에서, 도 10의 제2 무선 장치는 폴라 코드를 위한 파라미터(즉, K, N, M) 및 레이트 매칭 방식을 기반으로 정보 비트 집합, 프로즌 비트 집합, 펑처링 비트 집합 및 언노운 프로즌 비트 집합을 결정할 수 있다.
예를 들어 도 10의 크기 3의 정보 비트를 위한 최적화된 위치에 관한 정보 비트 집합(Ac1, 즉, u5, u6, u7)은 {5,6,7}일 수 있다.
일 예로, 도 6과 같이, 마더 행렬의 계층 특성에 따라 프로즌 비트 집합 및 펑처링 비트 집합의 인덱스는 {0,1,2,3}으로 동일할 수 있다.
예를 들어, 도 6과 같이, 언노운 프로즌 비트 집합의 인덱스는 존재하지 않을 수 있다.
S1050 단계에서, 도 10의 제2 무선 장치는 폴라 코드를 기반으로 복호화를 수행할 수 있다.
여기서, 도 10의 제2 무선 장치는 마더 행렬의 계층 특성을 기반으로 N보다 작은 크기의 폴라 코드로 복호화가 가능한지 여부를 판단할 수 있다.
예를 들어, 도 10의 제2 무선 장치는 도 6과 같이 부호화된 비트(예로, 도 6의 y4,y5,y6,y7)를 복호하기 위해, 길이 8의 마더 행렬(즉, 표 2)이 아닌 길이 4의 마더 행렬(즉, 표 1)을 이용할 수 있다.
S1060 단계에서, 도 10의 제2 무선 장치는 수신된 부호화된 비트에 대한 복호화 동작이 성공인지 여부를 판단할 수 있다.
만일 수신된 부호화된 비트에 대한 복호화가 성공으로 판단되면, S1070 단계에서, 도 10의 제2 무선 장치는 수신된 부호화된 비트로부터 길이 K의 정보 비트에 대한 정보를 획득할 수 있다.
만일 수신된 부호화된 비트에 대한 복호화가 실패로 판단되면, S1080 단계에서, 도 10의 제2 무선 장치는 재전송을 요청할 수 있다.
S1090 단계에서, 수신된 M개의 부호화된 비트들이 HARQ 재전송을 위해 수신된 것으로 판단되면, 도 10의 제2 무선 장치는 LLR 결합(Log-Likelihood Ratio combining)을 수행할 수 있다.
만일 HARQ 재전송을 위해 S1090 단계에 이어 S1050 단계가 수행될 때, 도 10의 언노운 프로즌 비트에 상응하는 부호화된 비트는 정보 복사 기법이 적용되어 먼저 부호화되는 위치의 비트 값을 이용하여 복호될 수 있다.
도 11은 본 일 실시 예에 따른 연속 펑처링(successive puncturing)의 치환된 천공 패턴(permuted punctiuring pattern)의 특성을 보여주는 개념도이다.
도 5 및 도 11을 참조하면, 입력 벡터(예로, u0, … u7)를 위한 위치 인덱스는 3비트의 이진수(b1,b2,b3)로 표현될 수 있다.
도 11에서, Πn은 길이 n의 모드 가능한 치환(permutation)을 포함하는 집합으로 정의될 수 있다. 예를 들어, Π3은 {(1,2,3), (1,3,2), (2,1,3), (2,3,1), (3,1,2), (3,2,1)}을 포함할 수 있다.
이 경우, 임의의 n에 대한 치환(permutation)인 σ가 정의될 수 있다. 여기서, σ는 Πn의 원소로 이해될 수 있다.
또한, 앞선 수학식 8의 시드 천공 패턴에 대한 비트-레벨 치환(bit-level permutation)이 수행된 천공 패턴의 위치 인덱스 집합은 하기의 수학식 9를 기반으로 정의될 수 있다.
본 명세서에서, 계층적 천공(Hierarchical Puncturing) 패턴은 수학식 8에 따른 시드 천공 패턴 및 수학식 9에 따른 비트-레벨 치환된 천공 패턴을 모두 포함하는 개념으로 이해될 수 있다.
예를 들어, 도 11의 천공 패턴(punctured coded bits)의 위치 인덱스는 B_PN={0,1,2,3}일 수 있다. 도 11을 참조하면, 비트-레벨 치환(bit-level permutation)를 위한 σ가 (3,2,1)일 때, 수학식 9에 따라 비트-레벨 치환된 천공 패턴의 위치 인덱스는 B_PN={0,4,2,6}일 수 있다.
다시 말해, 계층적 천공 패턴은 수학식 8에 따른 및 수학식 9에 따른 을 모두 포함할 수 있다. 계층적 천공 패턴을 사용될 때, 복호화 과정에서 짧은 길이(즉, )의 극 부호로 디코딩이 수행될 수 있다.
즉, 위 수학식 9에 따른 비트-레벨 치환(bit-level permutation)이 수행될 때, 폴라 코드의 마더 행렬의 구조적 특징에 따른 상호(reciprocal) 특성 및 계층(hierarchical) 특성이 그대로 유지됨은 이해될 것이다.
도 12 및 도 13은 본 다른 실시 예에 따라 HARQ를 지원하기 위해 폴라 코드를 기반으로 부호화하는 방법에 대한 개념도이다.
도 11 및 도 12를 참조하면, 길이 8의 폴라 코드로부터 출력되는 부호화된 비트의 길이가 '3'인 경우를 보여준다. 이 때, 도 12의 정보 비트 집합을 위한 최적화된 위치를 나타내는 인덱스 집합(A)은 {5,7}로 판단될 수 있다.
이어, 인코더의 역할을 수행하는 무선 장치는 인덱스 집합(A) 및 천공 패턴 {0,0,0,1,0,1,0,1}을 기반으로 부호화된 비트를 전송할 수 있다.
만일 도 12에서, 입력 벡터(u1, u3, u5, u7) 및 천공된 부호화된 비트(y3, y5, y7)를 이용하여 복호화 동작이 수행될 때, 계층적 펑처링의 특성이 고려되어, 길이 8의 마더 행렬(예로, 표 2)를 대신하여 길이 4의 마더 행렬(예로, 표 1)이 이용될 수 있음은 이해될 것이다.
도 11 내지 도 13을 참조하면, 재전송에 따라 길이 8의 폴라 코드로부터 출력되는 부호화된 비트의 길이가 '3'에서 '5'로 증가된 경우를 보여준다. 도 13의 경우, 도 13의 정보 비트 집합을 위한 최적화된 위치를 나타내는 인덱스 집합(A')은 {6,7}로 판단될 수 있다.
도 13과 같이 재전송이 수행될 때, 가변 부호율 조건을 만족시키기 위하여, 도 12의 위치 인덱스 집합(A)인 {5,7}에 상응하는 u5 및 u7에는 반드시 정보 비트가 포함될 필요가 있다.
도 13을 참조하면, u5에 있는 정보 비트는 정보 복사 기법에 따라 u6로 복사될 수 있다. 예를 들어, 디코딩 과정에서 u6가 u5보다 먼저 디코딩되는 위치라는 점이 고려될 수 있다. 또한, u5는 언노운 프로즌 비트로 사용될 수 있다.
결국, 도 13의 경우, 가변 부호율 조건을 만족하면서 동시에 인덱스 집합(A')을 통해 정보 비트가 전송될 수 있다.
이하, Hierarchical puncturing의 특징과 Information-copy기술을 이용한 가변 부호율을 지원하는 극 부호 설계 방법이 설명된다.
예를 들어, k 정보 비트를 m 개의 다양한 부호율(r1=k/N1 <...< rm=k/Nm)로 전송하는 가변 부호율 극 부호가 설계될 수 있다. 이 경우, m 개의 천공 패턴들은 p(1),p(2),...,p(m)으로 정의될 수 있다.
본 명세서에서, 천공패턴은 하기의 수학식 10과 같이 정의될 수 있다.
위 수학식 10에서, σ는 임의의 σ에 대하여 σ∈Πn를 만족하는 파라미터 값일 수 있다. 예를 들어, 가장 높은 부호율을 갖는 천공된 극 부호 (즉, 부호율 rm)를 이용하여 Information set A를 설계할 수 있다
위 주어진 천공 패턴들과 Information set A 및 하기 표 3의 방법을 기반으로 다른 천공된 극 부호들과 마더 극 부호에 적합한 Information set을 설계할 수 있다.
또한, 최적화된 Information set들과 하기 표 4의 알고리즘을 적용하여 non-zero frozen vector를 설계할 수 있다.
본 일 실시 예에 따른 가변 부호율 극 부호에 대한 인코딩 과정은 기존과 동일할 수 있다. 다만, all-zero frozen vector 대신에 하기 표 4를 통해 얻어진 non-zero frozen vector (또는 information-bit dependent frozen vector)를 이용할 수 있다. 예를 들어, information-copy 방법에 의해서 frozen vector의 일부분에는 정보 비트들이 복사될 수 있다.
도 13의 경우, 입력벡터는 (u0,u1,u2,u3,u4,u5,u6,u7)=(0,0,0,0,0,u5,u6,u7)일 수 있다. 즉, 논 제로 프로즌 벡터(non-zero frozen vector)는 (u0,u1,u2,u3,u4,u5,u6,u7)=(0,0,0,0,0,u5,0,0)일 수 있다. 다시 말해, 본 일 실시 예에 따른 frozen vector는 항상 all-zero vector가 아니고, information-bit에 의존하여 결정될 수 있다.
또한, 표 3의 알고리즘은 1회만 수행될 수 있으나, 하기의 표 4의 알고리즘은 non-zero frozen vector (or information-dependent frozen vector)를 구하기 위해서 HARQ-IR에서 매 초기 전송마다 수행될 수 있다.
도 14는 본 일 실시 예에 따른 가변 부호율을 지원하는 극 부호 설계에 대한 시뮬레이션 결과를 보여준다.
도 14를 참조하면, 정보 비트의 길이(K)는 '52'이고, K는 8비트의 CRC 비트를 포함한다고 가정할 수 있다. 또한, 부호율은 r1 (52/256) < r2 (52/192) < r3 (52/192) < r4 (52/64)이라고 가정할 수 있다. 이 경우, 천공 패턴은 하기위 수학식 11과 같이 결정될 수 있다.
예를 들어, HARQ-IR에서 첫 번째 전송에서, 부호율 r4인 가장 높은 부호율의 천공된 극 부호로 전송된다. 이어, 두 번째 재 전송에서, Bp(4)-Bp(3)에 속하는 인덱스들에 해당하는 리던던시 부호화 비트(redundancy coded bit)가 전송될 수 있다.
이어, 첫 번째와 두 번째 전송에서 받은 신호들이 결합되어 부호율 r3인 천공된 극 부호가 구성될 수 있다. 마찬가지로, 세 번째 전송에서는 Bp(3)-Bp(2)에 해당하는 리던던시 부호화 비트가 전송될 수 있다.
마지막으로, 네 번째 전송에서는 Bp(2)-Bp(1)에 해당하는 리던던시 부호화 비트가 전송될 수 있다. 즉, 네 번째 전송까지 받은 모든 수신 시그널이 결합되면 부호율 r1인 마더(mother) 극 부호가 구성될 수 있다.
도 14를 참조하면, 각 전송에 해당하는 부호율인 r4>r3>r2>r1에 해당하는 Frame-Error-Rate(FER)결과를 보여준다. 본 일 실시 예에 따르면, non-zero frozen vector (or information-dependent frozen vector)를 이용하여 family에 있는 각 천공된 극 부호에 대해서 최적화된 Information set이 사용될 수 있음이 보여진다.
도 15는 본 일 실시 예들이 적용될 수 있는 무선 장치들을 설명하기 위한 도면이다.
도 15의 기지국 장치(10)는, 수신 모듈(11), 송신 모듈(12), 프로세서(13), 메모리(14) 및 복수의 안테나(15)를 포함할 수 있다.
예를 들어, 수신 모듈(11)은 외부 장치(예를 들어, 단말)로부터 각종 신호, 데이터 및 정보를 수신할 수 있다. 송신 모듈(12)은 외부 장치(일 예로, 단말)로의 각종 신호, 데이터 및 정보를 전송할 수 있다. 이 경우, 수신 모듈(11)과 송신 모듈(12)은 하나의 트랜시버(transceiver)에 포함될 수 있다.
예를 들어, 프로세서(13)는 기지국 장치(10) 전반의 동작을 제어할 수 있다. 또한, 프로세서(13)는 기지국 장치(10)에 수신된 정보 및 기지국 장치(10)에 의해 외부로 송신될 정보를 연산 처리하는 기능을 수행할 수 있다.
예를 들어, 메모리(14)는 연산 처리된 정보를 소정 시간 동안 저장할 수 있다. 복수의 안테나(15)는 MIMO 송수신을 지원하기 위한 구성이며, 2차원 안테나 배치에 따라 구성될 수 있다.
도 15의 단말 장치(20)는 수신 모듈(21), 송신 모듈(22), 프로세서(23), 메모 리(24) 및 복수의 안테나(25)를 포함할 수 있다.
예를 들어, 수신 모듈(11)은 외부 장치(예를 들어, 단말)로부터 각종 신호, 데이터 및 정보를 수신할 수 있다. 송신 모듈(12)은 외부 장치(일 예로, 단말)로의 각종 신호, 데이터 및 정보를 전송할 수 있다. 이 경우, 수신 모듈(11)과 송신 모듈(12)은 하나의 트랜시버(transceiver)에 포함될 수 있다.
예를 들어, 프로세서(13)는 기지국 장치(10) 전반의 동작을 제어할 수 있다. 또한, 프로세서(13)는 기지국 장치(10)에 수신된 정보 및 기지국 장치(10)에 의해 외부로 송신될 정보를 연산 처리하는 기능을 수행할 수 있다.
예를 들어, 메모리(14)는 연산 처리된 정보를 소정 시간 동안 저장할 수 있다. 복수의 안테나(15)는 MIMO 송수신을 지원하기 위한 구성이며, 2차원 안테나 배치에 따라 구성될 수 있다.
본 명세서에서, 하향링크 전송 주체(entity) 또는 상향링크 수신 주체는 주로 기지국을 예로 설명되고, 하향링크 수신 주체 또는 상향링크 전송 주체는 주로 단말을 예로 설명된다. 다만, 본 명세서의 범위가 이에 제한 되는 것은 아님은 이해될 것이다.
본 명세서에서, 상술한 실시 예들은 다양한 수단을 통해 구현될 수 있다. 예를 들어, 상술한 실시 예들은 하드 웨어, 펌 웨어(firmware), 소프트 웨어 또는 그것들의 결합에 의해 구현될 수 있다.
하드 웨어에 의한 구현의 경우, 상술한 실시 예들은 하나 또는 그 이상의 ASICs(Application Specific Integrated Circuits), DSPs(Digital Signal Processors), DSPDs(Digital Signal Processing Devices), PLDs(Programmable Logic Devices), FPGAs(Field Programmable Gate Arrays), 프로세서, 컨트롤러, 마이크로 컨트롤러, 마이크로 프로세서 등에 의해 구현될 수 있다.
본 명세서의 상세한 설명에서는 구체적인 실시 예에 관하여 설명하였으나, 본 명세서의 범위에서 벗어나지 않는 한도 내에서 여러 가지 변형이 가능하다. 그러므로, 본 명세서의 범위는 상술한 실시 예에 국한되어 정해져서는 안되며 후술하는 특허청구범위뿐만 아니라 이 발명의 특허청구범위와 균등한 것들에 의해 정해져야 한다.
Claims (20)
- HARQ(Hybrid Automatic Repeat reQuest)를 지원하기 위해 폴라 코드를 기반으로 부호화하는 방법에 있어서,무선 장치가, 폴라 코드를 기반으로 미리 결정된 마더 행렬(mother matrix)에 적용되는 길이 N(N은 자연수)의 입력 벡터를 구성하되,상기 입력 벡터는 길이 K(K는 자연수)의 정보 비트의 최적화된 위치를 위한 제1 인덱스 집합, 상기 마더 행렬을 기반으로 생성되는 부호화된 비트 중 천공될 위치에 대한 제2 인덱스 집합 및 정보 복사(information copy) 기법과 연관된 적어도 하나의 언노운 프로즌 비트(unknown frozen bit)의 위치에 대한 제3 인덱스 집합을 기반으로 정의되는, 단계; 및상기 무선 장치가, 상기 입력 벡터 및 상기 마더 행렬을 기반으로 부호화를 수행하는 단계를 포함하는 방법.
- 제1 항에 있어서,상기 입력 벡터는 {u0,u1, …,uN-1}로 정의되고,상기 입력 벡터를 위한 위치 인덱스는 {0,1, …N-1}로 정의되고,상기 제1 인덱스 집합 내지 상기 제3 인덱스 집합은 상기 {0,1, …N-1}을 기반으로 정의되는 방법.
- 제2 항에 있어서,상기 N이 '8'일 때, 미리 결정된 위치 인덱스의 순서 {0}, {4}, {2}, {6}, {1}, {5}, {3}, {7}에 따라 디코딩되는 방법.
- 제2 항에 있어서,상기 입력 벡터는 이전의 전송을 위한 길이 I(I는 자연수)의 제2 정보 비트를 모두 포함하도록 설정되고, 상기 제2 인덱스 집합에 포함된 원소들의 개수는 상기 이전의 전송보다 감소되도록 설정되는 방법.
- 제4 항에 있어서,상기 K 및 I가 모두 '3'일 때, 상기 제1 인덱스 집합이 {a,b,c}이고, 상기 이전의 전송을 위한 상기 제2 정보 비트를 위한 최적화된 위치 인덱스가 {d,b,c}일 때, 상기 위치 인덱스 {d}에 포함된 정보는 상기 정보 복사 기법에 따라 상기 위치 인덱스 {d}보다 먼저 디코딩되는 위치 인덱스로 복사되는 방법.
- 제5 항에 있어서,상기 위치 인덱스 {d}가 위치 인덱스 {a}로 복사될 때, 상기 제3 인덱스 집합은 {d}를 포함하는 방법.
- 제7 항에 있어서,상기 마더 행렬의 크기는 N x N (N은 자연수)이고,상기 N은 2^n (n은 자연수)인 방법.
- 제1 항에 있어서,상기 제2 인덱스 집합에 상응하는 위치의 입력 벡터에는 미리 결정된 값을 갖는 프로즌 비트가 포함되는 방법.
- HARQ(Hybrid Automatic Repeat reQuest)를 지원하기 위해 폴라 코드를 기반으로 부호화하는 무선 장치에 있어서,무선 신호를 송신 및 수신하는 RF(Radio Frequency) 부; 및상기 RF부와 연결되는 프로세서(processor)를 포함하되, 상기 프로세서는,폴라 코드를 기반으로 미리 결정된 마더 행렬(mother matrix)에 적용되는 길이 N(N은 자연수)의 입력 벡터를 구성하도록 구현되고,상기 입력 벡터는 길이 K(K는 자연수)의 정보 비트의 최적화된 위치를 위한 제1 인덱스 집합, 상기 마더 행렬을 기반으로 생성되는 부호화된 비트 중 천공될 위치에 대한 제2 인덱스 집합 및 정보 복사(information copy) 기법과 연관된 적어도 하나의 언노운 프로즌 비트(unknown frozen bit)의 위치에 대한 제3 인덱스 집합을 기반으로 정의되고,상기 입력 벡터 및 상기 마더 행렬을 기반으로 부호화를 수행하도록 구현되는 무선 장치.
- 제11 항에 있어서,상기 입력 벡터는 {u0,u1, …,uN-1}로 정의되고,상기 입력 벡터을 위한 위치 인덱스는 {0,1, …N-1}로 정의되고,상기 제1 인덱스 집합 내지 상기 제3 인덱스 집합은 상기 {0,1, …N-1}을 기반으로 정의되는 무선 장치.
- 제12 항에 있어서,상기 N이 '8'일 때, 미리 결정된 위치 인덱스의 순서 {0}, {4}, {2}, {6}, {1}, {5}, {3}, {7}에 따라 디코딩되는 무선 장치.
- 제12 항에 있어서,상기 입력 벡터는 이전의 전송을 위한 길이 I(I는 자연수)의 제2 정보 비트를 모두 포함하도록 설정되고, 상기 제2 인덱스 집합에 포함된 원소들의 개수는 상기 이전의 전송보다 감소되도록 설정되는 무선 장치.
- 제14 항에 있어서,상기 K 및 I가 모두 '3'일 때, 상기 제1 인덱스 집합이 {a,b,c}이고, 상기 이전의 전송을 위한 상기 제2 정보 비트를 위한 최적화된 위치 인덱스가 {d,b,c}일 때, 상기 위치 인덱스 {d}에 포함된 정보는 상기 정보 복사 기법에 따라 상기 위치 인덱스 {d}보다 먼저 디코딩되는 위치 인덱스로 복사되는 무선 장치.
- 제15 항에 있어서,상기 위치 인덱스 {d}가 위치 인덱스 {a}로 복사될 때, 상기 제3 인덱스 집합은 {d}를 포함하는 무선 장치.
- 제17 항에 있어서,상기 마더 행렬의 크기는 N x N (N은 자연수)이고,상기 N은 2^n (n은 자연수)인 무선 장치.
- HARQ(Hybrid Automatic Repeat reQuest)를 지원하기 위해 폴라 코드를 기반으로 부호화된 비트를 전송하는 방법에 있어서,무선 장치가, 폴라 코드를 기반으로 미리 결정된 마더 행렬(mother matrix)에 적용되는 길이 N(N은 자연수)의 재전송을 위한 입력 벡터를 구성하되,상기 입력 벡터는 상기 마더 행렬을 기반으로 생성되는 부호화된 비트 중 천공될 위치에 대한 천공 인덱스 집합을 기반으로 정의되는, 단계; 및상기 무선 장치가, 상기 입력 벡터 및 상기 마더 행렬을 기반으로 부호화를 수행하는 단계;상기 무선 장치가, 상기 부호화를 수행하여 획득된 N 길이의 부호화된 비트 중 상기 천공 인덱스 집합에 따라 전송될 적어도 하나의 전송 비트를 판단하는 단계; 및상기 무선 장치가, 상기 적어도 하나의 전송 비트를 제2 무선 단말로 전송하는 단계를 포함하는 방법.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US16/756,069 US11239954B2 (en) | 2017-11-13 | 2018-11-13 | Encoding method on basis of polar code in order to support HARQ, and wireless device using same |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US201762585495P | 2017-11-13 | 2017-11-13 | |
| US62/585,495 | 2017-11-13 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2019093855A1 true WO2019093855A1 (ko) | 2019-05-16 |
Family
ID=66438454
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/KR2018/013793 Ceased WO2019093855A1 (ko) | 2017-11-13 | 2018-11-13 | Harq를 지원하기 위해 폴라 코드를 기반으로 부호화하는 방법 및 이를 이용한 무선 장치 |
Country Status (2)
| Country | Link |
|---|---|
| US (1) | US11239954B2 (ko) |
| WO (1) | WO2019093855A1 (ko) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019227276A1 (en) * | 2018-05-28 | 2019-12-05 | Qualcomm Incorporated | Polar code construction for incremental redundancy |
| US11050519B2 (en) * | 2018-09-14 | 2021-06-29 | Idac Holdings, Inc. | Methods, apparatus, systems and procedures for hybrid automatic repeat requests (HARQs) using polar codes |
| CN111447042B (zh) * | 2019-01-17 | 2021-12-24 | 华为技术有限公司 | 一种极化编译码方法及装置 |
| CN112889230A (zh) * | 2019-09-30 | 2021-06-01 | Oppo广东移动通信有限公司 | 上行控制信息的传输方法及装置 |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9742440B2 (en) * | 2015-03-25 | 2017-08-22 | Samsung Electronics Co., Ltd | HARQ rate-compatible polar codes for wireless channels |
| US9780918B2 (en) * | 2013-11-04 | 2017-10-03 | Huawei Technologies Co., Ltd. | Rate matching method and apparatus for polar codes, and wireless communication device |
| WO2017176309A1 (en) * | 2016-04-08 | 2017-10-12 | Intel Corporation | Polar codes for harq transmissions |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018227604A1 (en) * | 2017-06-16 | 2018-12-20 | Huawei Technologies Co., Ltd. | Methods and apparatus for polar encoding |
| US10700710B2 (en) * | 2017-07-24 | 2020-06-30 | Qualcomm Incorporated | Indicating a number of copied information bits in a retransmission |
| US11031958B2 (en) * | 2018-06-25 | 2021-06-08 | Qualcomm Incorporated | Hybrid polar code design for ultra-reliable low latency communications (URLLC) |
-
2018
- 2018-11-13 WO PCT/KR2018/013793 patent/WO2019093855A1/ko not_active Ceased
- 2018-11-13 US US16/756,069 patent/US11239954B2/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9780918B2 (en) * | 2013-11-04 | 2017-10-03 | Huawei Technologies Co., Ltd. | Rate matching method and apparatus for polar codes, and wireless communication device |
| US9742440B2 (en) * | 2015-03-25 | 2017-08-22 | Samsung Electronics Co., Ltd | HARQ rate-compatible polar codes for wireless channels |
| WO2017176309A1 (en) * | 2016-04-08 | 2017-10-12 | Intel Corporation | Polar codes for harq transmissions |
Non-Patent Citations (2)
| Title |
|---|
| HUAWEI ET AL.: "Summary of Polar Code Design for Control Channels", R1-1700088, 3GPP TSG RAN WG1 AD-HOC MEETING, 10 January 2017 (2017-01-10), Spokane, USA, XP051202597 * |
| KIM, JONG-HWAN ET AL.: "Low Complexity List Decoding for Polar Codes with Multiple CRC Codes", ENTROPY, vol. 19, no. 4, 24 April 2017 (2017-04-24), pages 183, XP055608263 * |
Also Published As
| Publication number | Publication date |
|---|---|
| US20200304241A1 (en) | 2020-09-24 |
| US11239954B2 (en) | 2022-02-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2018236114A1 (en) | METHOD AND APPARATUS FOR FLOW ADAPTATION FOR COMMUNICATION AND BROADCASTING SYSTEMS | |
| WO2018062660A1 (ko) | Qc ldpc 코드의 레이트 매칭 방법 및 이를 위한 장치 | |
| EP3625887A1 (en) | Method and apparatus of rate-matching for communication and broadcasting systems | |
| WO2017213409A1 (en) | Channel decoding method and apparatus using structured priori information of preamble | |
| WO2019190281A1 (en) | Apparatus and method for encoding and decoding using polar code in wireless communication system | |
| WO2015160205A1 (ko) | 부호어를 생성하는 방법 및 장치 그리고 부호어를 복원하는 방법 및 장치 | |
| WO2017217711A1 (ko) | 폴라 코드를 위한 데이터 재송신 방법 및 이를 위한 장치 | |
| WO2013032156A1 (en) | Method and apparatus for transmitting and receiving information in a broadcasting/communication system | |
| WO2019093855A1 (ko) | Harq를 지원하기 위해 폴라 코드를 기반으로 부호화하는 방법 및 이를 이용한 무선 장치 | |
| WO2009107989A2 (en) | Method and apparatus for channel encoding and decoding in a communication system using low-density parity-check codes | |
| WO2014092502A1 (en) | Encoding method and apparatus using crc code and polar | |
| WO2018128435A2 (ko) | 다중 ldpc 코드에서 ldpc 베이스 코드를 선택하는 방법 및 이를 위한 장치 | |
| WO2016089157A1 (en) | Sc-ldpc codes for wireless communication systems | |
| WO2018212558A1 (en) | Method and apparatus for coding/decoding in a communication or broadcasting system using high-order modulation | |
| WO2012093894A2 (en) | Method and apparatus for channel encoding and decoding in a broadcasting/communication system using low density parity-check codes | |
| WO2018143743A1 (ko) | 행-직교 구조(row-orthogonal)를 이용한 LDPC 코드 전송 방법 및 이를 위한 장치 | |
| WO2020067637A1 (ko) | 무선 통신 시스템에서 극 부호를 이용한 부호화 및 복호화를 위한 장치 및 방법 | |
| WO2017217715A1 (ko) | 폴라 코드를 위한 프로즌 비트 결정 방법 및 이를 위한 장치 | |
| WO2018079987A1 (ko) | Ldpc 코드의 운송블록 분할 방법 및 이를 위한 장치 | |
| WO2019031841A1 (ko) | 정보 전송 방법 및 전송 장치 및 정보 수신 방법 및 수신 장치 | |
| WO2019039742A1 (en) | METHODS AND APPARATUS FOR ENCODING AND DECODING CHANNELS USING ZERO BIT FILLING | |
| EP3963722A1 (en) | Apparatus and method for decoding of low-density parity check codes in wireless communication system | |
| WO2018226019A1 (ko) | 행-직교 구조를 이용한 ldpc 코드 전송 방법 및 이를 위한 장치 | |
| WO2018143717A1 (ko) | 폴라 코딩을 이용한 신호 송수신 방법 및 장치 | |
| WO2018084617A1 (ko) | Ldpc 코드의 운송블록 분할 방법 및 이를 위한 장치 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 18875677 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 18875677 Country of ref document: EP Kind code of ref document: A1 |