JP2008288681A - Equalization processing device, equalization processing method, and digital signal receiver - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an equalization processing device, an equalization processing method, and a digital signal receiver that can estimate transmission line characteristics using a control signal and correct waveform distortion caused by a multipath and a Doppler effect with higher precision. <P>SOLUTION: A signal sequence Xs comprising SP symbols transmitted as a reference signal at fixed intervals used for synchronization, distortion correction, etc., and TMCC symbols and AC symbols as a control signal arranged aperiodically is supplied to an LPF unit 134. At this time, a signal sequence Xp as position information representing carrier positions of the respective symbols is also supplied to the LPF unit 134, and interpolation in a frequency-axis direction is performed with the signal sequences Xs and Xp to estimate transmission line characteristics at the respective carrier positions. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an equalization processing device and an equalization processing method for performing equalization processing on a received digital communication signal, and a digital signal receiver that receives a digital signal including the equalization processing device.

  In recent years, with the development of digital compression coding technology and high-speed communication technology, digitization in satellite and terrestrial broadcast communication and digitization in mobile communication such as mobile phones has been realized. In addition, with the realization of digitalization in broadcast signals and mobile communication, this digital communication technology is used in various fields such as in-vehicle TV broadcast reception, portable mobile communication, wireless LAN, and Bluetooth.

  When digital communication technology is used for wireless communication, a digital signal receiver that receives a digital signal includes a direct wave that is received directly from a transmitter and a delayed wave that is reflected by an obstacle such as a building and received. Receive. For this reason, the received direct wave and the delayed wave interfere with each other, so that multipath distortion may occur in the received wave. Furthermore, in a digital signal receiver that performs mobile communication such as in-vehicle television broadcast reception and portable mobile communication, the received wave is also distorted by the Doppler effect (Doppler shift).

  In order to correct the distortion generated in the received wave in this way, for example, in an Orthogonal Frequency Division Multiplex (OFDM) transmission system, a pilot signal composed of known values is cyclic in the time direction and the frequency direction. Inserted. In the following, an ISDB-T (Integrated Services Digital Broadcasting-Terrestrial) system, which is a standard period of terrestrial digital television, will be described as a specific example.

  As shown in FIG. 11, the baseband signal constituting the OFDM signal according to the ISDB-T system is composed of data symbols and scattered pilot symbols (hereinafter referred to as “SP symbols”) arranged in the frequency direction and the time direction. These are collectively referred to as an OFDM symbol sequence. The frequency direction and the time direction are also called a carrier direction and a symbol direction, respectively.

  In FIG. 11, the time number (symbol number) corresponding to the time direction is represented by t (integer of t ≧ 0), and the carrier number corresponding to the frequency direction is l (0 ≦ l ≦ (L−1)). , L: total number of subcarriers). Further, t represents time when the symbol length of the OFDM signal is used as a unit. Further, a position in the OFDM symbol string that is uniquely determined by uniquely defining t and l is called a carrier position, and this carrier position is represented by (t, l).

  The SP symbol is arranged at a carrier position satisfying l = 3 × (t mod 4) + 12p. Note that mod represents a remainder operation, and p is an integer. That is, as shown in FIG. 11, when attention is paid to a signal at a certain time t, SP symbols are arranged on the frequency axis every 12 subcarriers. Each time the time t advances by one symbol, the SP signal is shifted in the frequency direction by three subcarriers. In other words, when attention is paid to a signal of a certain subcarrier, SP symbols are arranged every four symbols on the time axis. Thus, for example, at time t = 0, SP symbols are arranged at carrier positions (0, 0), (0, 12), (0, 24), (0, 36),. Then, SP symbols are arranged at carrier positions (1, 3), (1, 15), (1, 27), (1, 39),. Further, data symbols are arranged at carrier positions other than the carrier position where the SP symbols are arranged.

  With respect to the baseband signal having the configuration shown in FIG. 11, the transmission path characteristics for each subcarrier symbol, that is, for each carrier position, are estimated by SP symbols arranged in the frequency axis direction and the time axis direction. The equalization processing is performed based on the established transmission path characteristics. In this equalization processing, first, the transmission path characteristic with respect to the carrier position of each SP symbol is estimated by each SP symbol.

  At this time, for each subcarrier, interpolation in the time axis direction is performed using the transmission path characteristics for the carrier position of each SP symbol arranged on the same subcarrier every 4 symbols. That is, by performing such interpolation in the time axis direction, as shown in FIG. 11, in the symbols to be processed, carriers that are equally spaced between SP symbols (black circles in FIG. 11) arranged in advance. An interpolated SP symbol (white circle in FIG. 11) is arranged with respect to the position. Then, a zero value symbol is inserted into two carrier positions in each interval of each SP symbol that has been previously arranged or interpolated.

  In this way, the symbol to be processed for confirming the transmission path characteristic is a signal sequence composed of pre-arranged or interpolated SP symbols and zero value symbols, and low-pass filter processing is performed on this signal sequence. By this low-pass filter processing, the symbol to be processed is also subjected to interpolation in the frequency axis direction, and a waveform corresponding to that in which SP symbols are inserted into all carriers is obtained. As a result, the transmission path characteristics are estimated for all carrier positions in a symbol string having SP symbols every four symbols. By performing equalization processing using the channel characteristics estimated for each carrier position, the distortion of the baseband signal is corrected.

  However, when the delay time of a delayed wave due to multipath is long or when the moving speed in mobile communication is fast, the transmission path characteristics are violently converted in each of the time axis direction and the frequency direction. For this reason, the distortion generated in the received wave cannot be sufficiently corrected only by the above-described SP symbol. On the other hand, as a conventional technique, estimation of transmission path characteristics is made by referring to control signals such as TMCC (Transmission and Multiplexing Configuration Control) symbol, CP (Continual Pilot) symbol, AC (Auxiliary Channel) symbol and mobile body speed. A method for switching the method has been proposed (see Patent Document 1).

In the receiving apparatus of Patent Document 1, when the transmission path characteristics are estimated based on the amplitude fluctuation speed detected with respect to the control signals such as TMCC symbols, AC symbols, and CP symbols, or the amplitude fluctuation speed detected from the moving body speed. The estimation method used for is switched. That is, when the amplitude fluctuation speed is slow, as described above, after performing interpolation in the time axis direction (interpolation processing in the time axis direction) using the SP symbol, interpolation in the frequency axis direction (in the frequency axis direction) (Interpolation process) is performed (time interpolation + frequency interpolation method), and the channel characteristics are estimated from the SP symbols obtained for all carriers. Also, when the amplitude fluctuation speed is fast, interpolation for each carrier is performed by performing SP symbol interpolation or SP symbol and CP symbol extrapolation within the symbol to be processed in the frequency axis direction. (Symbol frequency interpolation method), and the transmission path characteristics are estimated from the SP symbols obtained for all carriers.
JP 2006-140987 A

  However, in the receiving apparatus of Patent Document 1, an SP symbol interpolation method for estimating transmission path characteristics is selected according to the amplitude fluctuation speed of the received signal. When the transmission path characteristics change more rapidly than the SP symbol period, the waveform distortion cannot be corrected sufficiently. That is, when waveform distortion occurs due to multipath and Doppler effects, such as when the delay time of a delayed wave due to multipath is long or when the moving speed in mobile communication is high, both in the time axis direction and frequency axis direction On the other hand, the transmission path characteristics change drastically. For this reason, even if any of the methods used in the receiving apparatus of Patent Document 1 is selected, the estimation accuracy of the transmission path characteristics is low for all carriers. Therefore, correction of waveform distortion by equalization processing cannot be made sufficient.

  In view of such a problem, the present invention provides an equalization processing apparatus capable of estimating transmission path characteristics using a control signal and correcting waveform distortion due to multipath and Doppler effects with higher accuracy. It is an object to provide a processing method and a digital signal receiver.

  In order to achieve the above object, the equalization processing apparatus of the present invention equalizes each channel of a digital signal that is transmitted by an orthogonal frequency division multiplexing system and that includes a reference signal that is periodically arranged at least in the frequency axis direction. In the equalization processing apparatus that performs processing, based on the transmission path characteristic estimation unit that estimates the transmission path characteristic of each channel in the frequency axis direction based on the reference signal, and the transmission path characteristic estimated by the transmission path characteristic estimation unit An equalization unit that performs equalization processing on each channel of the digital signal, and the digital signal includes a control signal that is continuously arranged aperiodically in the frequency axis direction, and the transmission path characteristics The estimation unit interpolates in the frequency axis direction a signal sequence composed of the reference signal and the control signal arranged in the frequency axis direction, thereby determining the channel characteristics of each channel in the frequency axis direction. Characterized in that it constant. The “channel” described above corresponds to a “subcarrier (carrier)” arranged in the frequency axis direction in the OFDM transmission method.

  In such an equalization processing apparatus, the transmission path characteristic estimation unit is configured by a filter for non-uniform sampling data, and performs a filtering process on a signal sequence based on the reference signal and the control signal by the filter. . Note that the filter may be constituted by a low-pass filter that passes a signal in a low frequency range.

  The filter further includes a first filter that performs a filtering process on a signal sequence based on the reference signal and the control signal, and a filter on a signal sequence based on positional information in the frequency axis direction of the reference signal and the control signal. A second filter that performs processing and a division circuit that divides the output of the first filter by the output of the second filter may be used. The first and second filters may be constituted by a low-pass filter that passes a signal in a low frequency range.

  Further, the reference signal is periodically arranged also in the time axis direction, and includes a time direction interpolation unit that interpolates the reference signal in the time axis direction, and a signal sequence based on the reference signal and the control signal includes: An interpolation reference signal obtained by interpolation in the time axis direction by the time direction interpolation unit may be included.

  In the above-described equalization processing device, pilot symbols such as SP (Scattered Pilot) symbols are used as the reference signal, and TMCC (Transmission and Multiplexing Configuration Control) symbols and AC (for example) are used as the control signals. Auxiliary Channel) symbols are used.

  The digital signal receiving apparatus of the present invention includes a fast Fourier transform unit that performs a fast Fourier transform on a digital signal transmitted by an orthogonal frequency division multiplexing method, and the digital signal that is converted into a signal in a frequency axis direction by the fast Fourier transform unit. An equalization processing unit that estimates transmission channel characteristics of each channel and performs equalization processing, and a digital modulation scheme in which the digital signal subjected to equalization processing by the equalization processing unit is assigned to each channel And a digital demodulator that demodulates the digital demodulator according to the present invention, wherein the equalization processor is configured by any of the above-described equalization processors.

  The equalization processing method of the present invention is an equalization processing method for performing equalization processing on each channel of a digital signal that is transmitted by an orthogonal frequency division multiplexing method and that includes a reference signal periodically arranged at least in the frequency axis direction. A channel characteristic estimation step for estimating a channel characteristic of each channel in the frequency axis direction based on the reference signal, and each channel of the digital signal based on the channel characteristic estimated in the channel characteristic estimation step An equalization step for performing equalization processing, and the digital signal includes a control signal that is continuously arranged aperiodically in the frequency axis direction, and in the transmission path characteristic estimation step, By interpolating in the frequency axis direction a signal string composed of the reference signal and the control signal arranged side by side, the transmission path characteristics of each channel in the frequency axis direction are obtained. And estimating a.

  According to the present invention, the channel characteristics in each channel can be estimated by performing interpolation in the frequency axis direction using not only an existing reference signal but also a control signal added by an existing standard. As a result, the number of signals used for distortion correction in the equalization process can be increased as compared with the conventional case, so that distortion correction accuracy can be increased. Therefore, sufficiently accurate distortion correction can be performed even when the delay time of delayed waves due to multipath is long or when the moving speed in mobile communication is high.

  First, the configuration of a digital signal receiving apparatus according to an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing an internal configuration of a digital signal receiving apparatus according to this embodiment. In the present embodiment, the above-described ISDB-T terrestrial digital television broadcast receiving apparatus will be described as an example.

  The digital signal receiving apparatus shown in FIG. 1 includes an antenna 10 that receives a digital broadcast signal, and an RF (Radio Frequency) signal processing unit 11 that selects a digital broadcast signal in a desired band from the antenna 10 and converts it into a baseband signal. And an FFT unit 12 that performs Fast Fourier Transform (FFT) on the baseband signal obtained by channel selection by the RF signal processing unit 11, and an FFT process performed by the FFT unit 12 to obtain a frequency from a time-axis signal. An equalization processing unit 13 for performing an equalization process on the baseband signal converted into the axial signal, and a baseband signal from which multipath distortion received during propagation by the equalization processing unit 13 is removed is digitally modulated. A digital demodulator 14 that demodulates based on the MPEG, and an MPEG (Moving Picture Experts Group) encoded signal demodulated by the digital demodulator 14 as an MPEG compression method It comprises an MPEG decoder 15 for decoding based on the.

  This receiving apparatus for receiving a digital broadcast signal receives a digital broadcast by the OFDM transmission method from an antenna 10. This OFDM scheme is a scheme in which a large number of subcarriers orthogonal to each other are multiplexed and transmitted within one channel band. At this time, the RF signal processing unit 11 selects an OFDM signal (digital broadcast signal) that is a high-frequency signal of a desired channel, and converts the selected OFDM signal to a baseband signal that is an IF (Intermediate Frequency) signal. (Down-convert). When the obtained baseband signal is given to the FFT unit 12, the baseband signal that is a time axis signal is converted into a baseband signal that is a frequency axis signal by FFT processing.

  When the FFT unit 12 performs FFT processing, symbol synchronization, carrier frequency synchronization, and sampling frequency synchronization are established. That is, the offset of the symbol length (time length of each symbol), the carrier frequency, and the sampling frequency (frequency of the subcarrier interval) generated by a transmission path such as multipath is corrected. At this time, symbol synchronization and carrier frequency synchronization are established by using a guard interval provided for each symbol and comprising the same signal as the latter half of the symbol. Also, sampling frequency synchronization is established by estimating the frequency offset using pilot symbols described later. The carrier frequency synchronization may be established by estimating a frequency offset using a pilot symbol.

  As described above, the baseband signal FFT-processed by the FFT unit 12 includes data symbols and pilot symbols arranged in the frequency direction and the time direction (in the ISDB-T system in this embodiment, “SP Symbol ") (see FIG. 11). Furthermore, in this embodiment, a TMCC symbol and an AC symbol (white triangles in FIG. 2) are further added as control signals to the OFDM symbol sequence composed of the data symbols and SP symbols.

  These TMCC symbols and AC symbols are control signals added to transmit control information such as modulation scheme and interleave and additional information, and are not periodically arranged in the frequency axis direction like SP symbols. . That is, as shown in FIG. 2, similarly to the case of FIG. 11, the TMCC symbol and the AC symbol are non-periodic with respect to the SP symbol arranged at the carrier position satisfying l = 3 × (t mod 4) + 12p. At random (random) carrier positions. The TMCC symbols and AC symbols are continuously arranged in the time axis direction.

  In this way, when the baseband signal has a configuration in which the TMCC symbol and the AC symbol are added to the OFDM symbol sequence of the data symbol and the SP symbol, the equalization processing unit to which the baseband signal from the FFT unit 12 is given 13, first, the transmission path characteristics for each carrier position are estimated using the SP symbol, the TMCC symbol, and the AC symbol. At this time, low-pass processing is performed in the frequency axis direction in order to estimate transmission path characteristics. Transmission path characteristic estimation processing combined with this low-pass processing will be described later.

  As described above, when the channel characteristics of all data symbols are estimated based on the SP symbols, the estimated channel characteristics are complex for each data symbol obtained from the frequency axis signal from the FFT unit 12. By the division, equalization processing is performed, and distortions in amplitude and phase due to multipath effects and the like are removed. As this equalization processing, for example, a zero focusing equalization method that directly divides the estimated transmission path characteristic is used. Since this zero-focusing equalization process has a problem of noise enhancement, there is a Minimum Mean Square Error (MMSE) equalization method for reducing the noise enhancement problem. In the case of this MMSE equalization method, it is also necessary to estimate the average power value of noise (additional noise) added on the transmission line.

  In this way, the baseband signal that has been equalized for each symbol is demodulated by the digital demodulation unit 14 using the digital modulation scheme set for each subcarrier. Examples of the digital modulation method include QAM (Quadrature Amplitude Modulation), QPSK (Quadrature Phase Shift Keying), and DQPSK (Differential Quadrature Phase Shift Keying). Then, when the MPEG encoded signal obtained by demodulating by the digital demodulator 14 is given to the MPEG decoder 15, it is decoded based on the MPEG compression method to obtain a video signal. This video signal is further applied to a display device (not shown) such as a display, so that the video is reproduced.

  A detailed example of the equalization processing unit 13 in such a digital signal receiving apparatus will be described below.

(First example of equalization processing unit)
A first example of the equalization processing unit 13 in the digital signal receiving apparatus configured as shown in FIG. 1 will be described with reference to the drawings. FIG. 3 is a block diagram showing the internal configuration of the equalization processing unit in this example.

  As shown in FIG. 3, the equalization processing unit 13 of the present example includes an SP extraction unit 131 that extracts an SP symbol from a baseband signal that has been FFT processed by the FFT unit 12, and a baseband that has been FFT processed by the FFT unit 12. A TMCC / AC extraction unit 132 that extracts TMCC symbols and AC symbols from the signal, and a positional information generation unit 133 that stores the positional information of TMCC symbols and AC symbols on the frequency axis and outputs the combined positional information of SP symbols. A low pass filter (LPF) unit 134 that outputs the channel characteristics given the SP symbol, the TMCC symbol, and the AC symbol obtained by the SP extraction unit 131 and the TMCC / AC extraction unit 132, and the LPF unit 134. Baseband signal etc. based on transmission path characteristics obtained by filtering Includes an equalizer 135 that processing performed, the.

  In such a configuration, since the baseband signal based on the OFDM symbol sequence as shown in FIG. 2 is provided from the FFT unit 12, the SP extraction unit 131 has l = 3 × (t mod 4) for each symbol sequence. SP symbols arranged at carrier positions satisfying + 12p are extracted. At this time, about 489 SP symbols are extracted from one symbol string composed of 5617 carriers. Also, TMCC / AC extraction section 132 extracts TMCC symbols and AC symbols at the same carrier position in each symbol string. The total number of TMCC symbols and AC symbols is 156 in one symbol string.

  Then, the SP symbol extracted by the SP extraction unit 131 and the TMCC symbol and the AC symbol extracted by the TMCC / AC extraction unit 132 are provided to the LPF unit 134. Therefore, as shown in FIG. 4 (a), for carrier positions without SP symbols (black circles in FIG. 4 (a)), TMCC symbols and AC symbols (white triangles in FIG. 4 (a)), By inserting a zero value, the signal sequence Xs of 5617 points is given to the LPF unit 134. As shown in FIG. 4A, the signal sequence Xs is a signal sequence in the frequency axis direction. In the equalization processing unit 13, values for each frequency axis are processed in time series.

  In addition, the position information generation unit 133 generates position information indicating the carrier position of each of the SP symbol, TMCC symbol, and AC symbol in the symbol sequence processed by the equalization processing unit 13, and sends it to the LPF unit 134. The That is, as shown in FIG. 4B, the signal sequence Xp constituting the position information sent to the LPF unit 134 is at the carrier position where any of the SP symbol, TMCC symbol, and AC symbol is arranged. The value is 1, and the value at the carrier position where none of the SP symbol, TMCC symbol, and AC symbol is arranged is 0. This signal sequence Xp is also a signal sequence in the frequency axis direction as shown in FIG. 4B, but the value for each frequency axis is processed in time series in the equalization processing unit 13.

  In this way, the LPF unit 134 to which the signal sequences Xs and Xp are given performs a filtering process to be described later, so that the frequency axis is based on the signal values of the carrier positions of the SP symbol, TMCC symbol, and AC symbol of the signal sequence Xs. Directional interpolation is performed. Thereby, the transmission path characteristics at all carrier positions of the symbol sequence to be processed are estimated and supplied to the equalizer 135. As described above, the equalizer 135 performs an equalization process on the baseband signal subjected to the FFT process by the FFT unit 12 by performing complex division on the estimated transmission path characteristics, thereby performing multipath. The distortion in the amplitude and phase due to the influence of the above is removed.

(1) Overview of Low-Pass Filter Processing An overview of the low-pass filter processing in the LPF unit 134 to which the above-described signal sequences Xs and Xp are input will be described below with reference to the drawings. In the following, a description will be given in comparison with a case where low-pass filter processing is performed using only SP symbols. In addition, a signal sequence using only SP symbols is Xs1, and the values for each frequency axis are processed in time series in the same manner as the signal sequences Xs and Xp described above.

  As shown in FIG. 5 (a), the signal sequence Xs1 is composed of SP symbols that are periodic in the frequency axis direction. Therefore, the signal sequence Xs1 in which the signal value based on the SP symbols appears every predetermined time (FIG. 5 (a) When low-pass filter processing is performed on the solid line A), an envelope (dotted line B in FIG. 5A) connecting signal values based on SP symbols is detected. The low-pass filter processing of this signal sequence Xs1 will be described in more detail.

  That is, when the signal sequence Xs1 is converted into a time-series signal sequence, when the signal sequence Xs1 is frequency-converted, as shown in FIG. 5B, a high-frequency component H based on a periodic SP symbol is detected. A low-frequency component L representing the obtained envelope is acquired. At this time, the high-frequency component H represents the periodicity of the periodic SP symbol on the frequency axis, and thus has a narrow bandwidth. Therefore, as shown in FIG. 5B, if a component having a frequency f or less can pass through the low-pass filter process, a low-frequency component L representing an envelope can be extracted.

  However, as described in [Background Art], the low-pass filter processing is performed on the signal sequence Xs1 including only SP symbols when the delay time of the delayed wave due to multipath is long or when the moving speed is high in mobile communication. I can't follow up enough. Therefore, as described above, the signal sequence Xs to which the aperiodic TMCC symbol and the AC symbol are added as shown in FIG. 6A is used. That is, the signal sequence Xs is a low-frequency signal generated by an envelope that connects a high-frequency component (solid line A1 in FIG. 6A) based on the SP symbol, TMCC symbol, and AC symbol and a value based on the SP symbol, TMCC symbol, and AC symbol. Component (dotted line B1 in FIG. 6A).

  Therefore, when the signal sequence Xs is converted into a time-series signal sequence, when the signal sequence Xs is frequency-converted, as shown in FIG. 6B, the high-frequency component H1 based on the SP symbol, the TMCC symbol, and the AC symbol, The low frequency component L1 representing the envelope obtained by detection is acquired. At this time, the high frequency component H1 represents the aperiodicity of each signal position on the frequency axis in which the aperiodic TMCC symbol and the AC symbol are added for each periodic SP symbol. Compared with the high frequency component H1 (see FIG. 5B), the bandwidth is widened. Therefore, as shown in FIG. 6B, when a component having a frequency of f or less can pass through the low-pass filter process, not only the low-frequency component L1 representing the envelope but also a part h1 of the high-frequency component H1 is extracted. End up.

  With respect to this signal sequence Xs, as shown in FIG. 7A, the signal sequence Xp has the SP symbol, TMCC symbol, and AC symbol signal values of 1, so that the SP symbol, TMCC symbol, and AC symbol It is comprised by the high frequency component by this signal value. That is, when the signal sequence Xp is converted into a time-series signal sequence, when the signal sequence Xp is frequency-converted, as shown in FIG. 7B, the high frequency component H1 due to the SP symbol, the TMCC symbol, and the AC symbol is obtained. To be acquired. The high frequency component H1 is the same as the high frequency component H1 (see FIG. 6B) acquired from the signal sequence Xs.

  Therefore, as shown in FIG. 7B, when a component having a frequency of f or less can pass through the low-pass filter process, a part h1 of the high-frequency component H1 is extracted. Therefore, by removing the frequency component h1 of the low-pass filtered signal sequence Xp from the frequency components L1 and h1 of the low-pass filtered signal sequence Xs, the low-frequency component L1 representing the envelope can be obtained. Thereby, interpolation in the frequency axis direction is performed by the signal sequence Xs, and transmission path characteristics at all carrier positions of the symbol sequence to be processed are estimated.

That is, assuming that the signal sequences Xs and Xp are functions Xs (t) and Xp (t) that change over time, and x (t) is a data amount that changes over time, the function Xs (t ) And Xp (t) are expressed by the following equation (1). That is, the signal sequence Xs (t) is obtained by sampling at a discontinuous timing represented by Xp (t) with respect to x (t) whose data amount continuously changes. .
Xs (t) = x (t) × Xp (t) (1)

When such a signal sequence Xs (t) is subjected to a low-pass filter process, it is approximately expressed as equation (2). Xs _LP and Xp _LP are functions after low-pass filter processing of the signal sequences Xs and Xp, respectively.
Xs _LP (t) ≈x (t) × Xp _LP (t) (2)

Then, as in the equation (3), the signal sequence Xs _LP (t) after the low-pass filter processing is divided by the signal sequence Xp _LP (t) after the low-pass filter processing, so that it continuously changes before sampling. x (t) can be obtained approximately. Thereby, interpolation in the frequency axis direction is performed on the signal sequence Xs to estimate the transmission path characteristics at all carrier positions of the symbol sequence to be processed.
x (t) ≈Xs _LP (t) / Xp _LP (t) (3)

  As described above, the signal sequence Xs is a signal sequence to which a non-periodic TMCC symbol and an AC symbol are added, unlike the case of processing the signal sequence Xs1 based on the periodic SP symbol. A process for non-uniform sampling data is performed. That is, unlike the processing for the signal sequence Xs1, which is uniform sampling data, low-pass filter processing is performed using the signal sequence Xp at the sampling position corresponding to the carrier position of each symbol in the signal sequence Xs.

(2) Configuration of LPF As described above, the LPF unit 134 first generates a signal sequence Xs based on the SP symbols extracted by the SP extraction unit 131 and the TMCC symbols and AC symbols extracted by the TMCC / AC extraction unit 132. Low-pass filter processing is performed, and the signal sequence Xp based on position information given from the position information generation unit 133 is low-pass filtered. Then, the signal sequence Xs after the low-pass filter processing is divided by the signal sequence Xp after the low-pass filter processing to obtain a signal sequence interpolated in the frequency axis direction with respect to the signal sequence Xs, and the symbol sequence to be processed The channel characteristics at all carrier positions are estimated.

  That is, as shown in FIG. 8, the LPF unit 134 includes an LPF 41 to which a signal sequence Xs is provided by the SP symbol extracted by the SP extraction unit 131 and the TMCC symbol and the AC symbol extracted by the TMCC / AC extraction unit 132. The LPF 42 to which the signal sequence Xp based on the position information from the position information generation unit 133 is given, and the division circuit 43 that divides the output from the LPF 41 by the output from the LPF 42. For example, FIR (Finite Impulse Response) type low-pass filters can be used as the LPFs 41 and 42 used in the LPF unit 134 configured as shown in FIG.

With this configuration, the SP symbol extracted by the SP extraction unit 131 and the TMCC symbol and the AC symbol extracted by the TMCC / AC extraction unit 132 are given to the LPF 41. When no SP symbol, TMCC symbol, or AC symbol is input, a zero value is inserted. By operating in this way, the signal sequence Xs as shown in FIG. 4A is input to the LPF 41, low-pass filter processing is performed on the signal sequence Xs, and the signal sequence Xs _LP after low-pass filter processing is divided into the divider circuit 43. Is output.

Further, a signal sequence Xp as shown in FIG. 4B is given to the LPF 42 from the position information generation unit 133. Therefore, the LPF 42 performs low-pass filter processing on the signal sequence Xp, and the signal sequence Xp _LP after low-pass filter processing is output to the division circuit 43. Then, the division circuit 43 divides the signal sequence Xs _LP by the signal sequence Xp _LP to obtain the signal sequence x obtained by performing interpolation in the frequency axis direction using the signal sequence Xs, and an equalizer as the estimated transmission line characteristic It outputs to 135.

(Second example of equalization processing unit)
A second example of the equalization processing unit 13 in the digital signal receiving apparatus configured as shown in FIG. 1 will be described with reference to the drawings. FIG. 9 is a block diagram showing the internal configuration of the equalization processing unit in this example. In the configuration of FIG. 9, portions used for the same purpose as the configuration of FIG. 3 are denoted by the same reference numerals, and detailed description thereof is omitted.

  As shown in FIG. 9, the equalization processing unit 13 of this example further performs interpolation in the time axis direction on the SP symbols extracted by the SP extraction unit 131 in addition to the configuration shown in FIG. This is a configuration in which an SP time direction interpolation unit 136 for giving a symbol to the LPF unit 134 is added. The SP time direction interpolation unit 136 uses, for example, an IIR (Infinite Impulse Response) type low-pass filter or the like to perform interpolation processing in the time axis direction from SP symbols arranged every four symbols at the same carrier position. I do. Thus, three interpolated SP symbols are arranged for every three carriers between the SP symbols in the symbol sequence to be processed.

  Therefore, SP symbols (black dots in FIG. 10A) arranged every 12 carriers are given to the SP time direction interpolation unit 136, so that every 3 carriers between SP symbols adjacent in the frequency axis direction. Interpolation is performed using SP symbols adjacent in the time axis direction. As a result, the SP time direction interpolation unit 136 interpolates SP symbols (white circles in FIG. 10A) every three carriers between SP symbols adjacent in the frequency axis direction, as shown in FIG. 10A. Is obtained and given to the LPF unit 134.

  Then, the SP symbol extracted by the SP extraction unit 131, the TMCC symbol and the AC symbol extracted by the TMCC / AC extraction unit 132, and the interpolation SP symbol acquired by the SP time direction interpolation unit 136 are combined into the LPF unit 134. Given to. Therefore, as shown in FIG. 10A, SP symbols (black circles in FIG. 10A), interpolation SP symbols (white circles in FIG. 10A), TMCC symbols, and AC symbols (FIG. 10A). For a carrier position having no white triangle in the middle, a signal value Xs of 5617 points is given to the LPF unit 134 by inserting a zero value.

  Further, the position information generation unit 133 generates position information indicating the carrier positions of the SP symbol, the interpolation SP symbol, the TMCC symbol, and the AC symbol in the symbol sequence processed by the equalization processing unit 13, and the LPF unit 134. Note that, unlike the first example, this position information generation unit 133 also indicates position information on the carrier position of the interpolated SP symbol, so that the same position information is obtained at any symbol position.

  That is, the position information generation unit 133 stores position information indicating the carrier positions of the SP symbol and the interpolation SP symbol for every three carriers, and the carrier position of the TMCC symbol and the AC symbol, and the stored position information is stored in the LPF unit. 134. As shown in FIG. 10B, the signal sequence Xp constituting the position information given to the LPF unit 134 is a carrier in which any of SP symbols, interpolation SP symbols, TMCC symbols, and AC symbols are arranged. The value at the position is 1, and the value at the carrier position where none of the SP symbol, the interpolated SP symbol, the TMCC symbol, and the AC symbol is arranged is 0.

  Then, the LPF unit 134 performs low-pass filter processing on each of the above-described signal sequences Xs and Xp. When the LPF unit 134 is configured as shown in FIG. 8 as in the first example, a signal string Xs as shown in FIG. 10A is given to the LPF 41 and a signal string as shown in FIG. Xp is given to the LPF 42. At this time, the SPF extracted by the SP extraction unit 131, the TMCC symbol and the AC symbol extracted by the TMCC / AC extraction unit 132, and the interpolation SP symbol acquired by the SP time direction interpolation unit 136 with respect to the LPF 41. And is given. When no SP symbol, interpolated SP symbol, TMCC symbol, or AC symbol is input, a zero value is inserted.

Then, in the LPF 41 to which the signal sequence Xs as shown in FIG. 10A is input, the low-pass filter processing is performed on the signal sequence Xs, and the signal sequence Xs _LP after the low-pass filter processing is output to the division circuit 43. Further, in the LPF 42 to which the signal sequence Xp as shown in FIG. 10B is input, the low-pass filter processing is performed on the signal sequence Xp, and the signal sequence Xp _LP after the low-pass filter processing is output to the division circuit 43. Then, the division circuit 43 divides the signal sequence Xs _LP by the signal sequence Xp _LP to obtain the signal sequence x obtained by performing interpolation in the frequency axis direction using the signal sequence Xs, and an equalizer as the estimated transmission line characteristic It outputs to 135.

  As described above, in the present embodiment, the signal used for estimating the transmission path characteristics is used for synchronization or distortion correction of pilot symbols (including SP signals) as in the conventional case, so that a constant interval is used. In addition to the reference signal transmitted in the above, control signals such as TMCC symbols and AC symbols are additionally used. As a result, the number of signals used for estimating the transmission path characteristics increases, so that the estimation accuracy can be increased, and the waveform distortion correction accuracy by the equalization processing can be increased.

  INDUSTRIAL APPLICABILITY The present invention is applicable to a digital signal receiving apparatus that receives a digital signal transmitted by a digital communication system having a reference signal transmitted at regular intervals for use in synchronization of pilot symbols and the like, distortion correction, and the like. . That is, the present invention can be applied to a digital television receiver that receives digital television broadcasts, a digital signal receiver that includes a mobile phone that performs digital mobile communication, and the like.

These are block diagrams which show the internal structure of the digital signal receiver in embodiment of this invention. FIG. 4 is a diagram illustrating a relationship between symbols including a TMCC symbol and an AC symbol in an OFDM symbol sequence. These are block diagrams which show the 1st example of a structure of the equalization process part in the digital signal receiver of FIG. These are figures which show the structure of the signal sequence processed in the equalization process part of FIG. These are figures for demonstrating the low-pass filter process in the signal sequence Xs1 only by SP symbol. These are the figures for demonstrating the low-pass filter process in the signal sequence Xs by SP symbol, TMCC symbol, and AC symbol. These are the figures for demonstrating the low-pass filter process in the signal sequence Xp which shows the carrier position of SP symbol, TMCC symbol, and AC symbol. FIG. 3 is a block diagram showing a configuration of an LPF unit. These are block diagrams which show the 2nd example of a structure of the equalization process part in the digital signal receiver of FIG. These are figures which show the structure of the signal sequence processed in the equalization process part of FIG. FIG. 4 is a diagram showing a relationship between SP symbols and data symbols in an OFDM symbol sequence.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Antenna 11 RF signal processing part 12 FFT part 13 Equalization processing part 14 Digital demodulation part 15 MPEG decoder 131 SP extraction part 132 TMCC / AC extraction part 133 Position information generation part 134 LPF part 135 Equalizer 136 SP Time direction interpolation part 41, 42 LPF
43. Division circuit

Claims (6)

In an equalization processing apparatus for performing equalization processing on each channel of a digital signal including a reference signal transmitted by an orthogonal frequency division multiplexing method and periodically arranged at least in the frequency axis direction,
A transmission path characteristic estimation unit that estimates transmission path characteristics of each channel in the frequency axis direction based on the reference signal;
An equalization unit that performs equalization processing on each channel of the digital signal based on the transmission path characteristic estimated by the transmission path characteristic estimation unit;
With
The digital signal comprises a control signal that is continuously arranged aperiodically in the frequency axis direction,
The transmission path characteristic estimation unit estimates a transmission path characteristic of each channel in the frequency axis direction by interpolating in the frequency axis direction a signal sequence composed of the reference signal and the control signal arranged in the frequency axis direction. An equalization processing apparatus characterized by. The transmission path characteristic estimation unit is configured by a filter for non-uniform sampling data,
The equalization processing apparatus according to claim 1, wherein a filter process is performed on a signal sequence based on the reference signal and the control signal by the filter. The filter is
A first filter that performs a filtering process on a signal sequence based on the reference signal and the control signal;
A second filter that performs a filtering process on a signal sequence based on positional information in the frequency axis direction of the reference signal and the control signal;
A division circuit for dividing the output of the first filter by the output of the second filter;
The equalization processing apparatus according to claim 2, comprising: The reference signal is periodically arranged also in the time axis direction,
A time direction interpolation unit for interpolating the reference signal in the time axis direction;
4. The signal sequence based on the reference signal and the control signal includes an interpolation reference signal obtained by interpolation in the time axis direction by the time direction interpolation unit. The equalization processing apparatus described in 1. A fast Fourier transform unit that performs fast Fourier transform on a digital signal transmitted by an orthogonal frequency division multiplexing system, and transmission channel characteristics of each channel with respect to the digital signal converted into a signal in the frequency axis direction by the fast Fourier transform unit An equalization processing unit that estimates and performs equalization processing, and a digital demodulation unit that demodulates the digital signal subjected to equalization processing by the equalization processing unit using a digital modulation scheme assigned to each channel, In the digital signal receiving device provided,
The digital signal receiving apparatus, wherein the equalization processing unit is configured by the equalization processing apparatus according to any one of claims 1 to 4. In an equalization processing method for performing equalization processing on each channel of a digital signal including a reference signal that is transmitted by an orthogonal frequency division multiplexing method and periodically arranged in the frequency axis direction,
A channel characteristic estimation step for estimating a channel characteristic of each channel in the frequency axis direction based on the reference signal;
An equalization step of performing equalization processing on each channel of the digital signal based on the transmission path characteristics estimated in the transmission path characteristics estimation step;
With
The digital signal comprises a control signal that is continuously arranged aperiodically in the frequency axis direction,
In the transmission line characteristic estimation step, the transmission line characteristic of each channel in the frequency axis direction is estimated by interpolating in the frequency axis direction a signal sequence composed of the reference signal and the control signal arranged in the frequency axis direction. An equalization processing method characterized by the above.

Images