JP5395617B2 - Receiving apparatus and receiving method for repeatedly decoding concatenated code data - Google Patents

Description

  The present invention relates to a reception technique for repeatedly decoding outer code and inner code concatenated code data in a transmission system for transmitting and receiving digital data such as terrestrial digital television broadcast and terrestrial digital audio broadcast.

  Conventionally, in terrestrial digital broadcasting (ISDB-T (Integrated Services Digital Broadcasting for Terrestrial)) of terrestrial digital television broadcasting and terrestrial digital audio broadcasting, external codes such as RS (Reed Solomon) codes and convolutional codes are used. By using a concatenated code combined with an inner code, an error occurring on the propagation path is corrected.

  First, a transmission apparatus used in a terrestrial digital broadcast transmission system will be described. FIG. 14 is a block diagram illustrating a configuration of the transmission apparatus. This transmission apparatus 100 includes an RS encoding unit 71, an energy spreading unit 72, a byte interleaving unit 73, a convolutional encoding unit 74, a mapping unit 75, a time interleaving unit 76, a frequency interleaving unit 77, an OFDM (Orthogonal Frequency Division Multiplexing: orthogonal frequency division). A multiplex (frame) unit 78, an IFFT (Inverse Fast Fourier Transform) unit 79, and a GI (Guard Interval) guard unit 80. The transmission apparatus 100 performs processing such as RS coding and convolutional coding on video / audio data of terrestrial digital broadcasting, and transmits an OFDM modulated wave of concatenated code data from a transmission antenna.

  The terrestrial digital broadcast video / audio data is outer-coded by the RS code unit 71 using the RS code, subjected to energy diffusion processing by the energy diffusion unit 72, byte interleaved by the byte interleaving unit 73, and convolutional code unit. In 74, the inner code is encoded by the convolutional code, and is mapped by the mapping unit 75 into the I and Q signals (In-phase: In-phase component signal, Quadrature-phase: Quadrature component signal).

  The I and Q signals mapped by the mapping unit 75 are time interleaved by the time interleaving unit 76, frequency interleaved by the frequency interleaving unit 77, and configured into an OFDM frame by the OFDM framing unit 78.

  The I and Q signals configured in the OFDM frame by the OFDM framing unit 78 are subjected to inverse fast Fourier transform from the frequency domain signal to the time domain signal by the IFFT unit 79, and the GI adding unit 80 adds the GI. . The video / audio data of the terrestrial digital broadcast processed in this way is transmitted from the transmission antenna as an OFDM modulated wave.

  Next, a conventional receiving apparatus used in a terrestrial digital broadcast transmission system will be described. FIG. 15 is a block diagram showing a configuration of a conventional receiving apparatus. The receiving apparatus 101 includes a symbol synchronization unit 11, a GI removal unit 12, an FFT (Fast Fourier Transform) unit 13, a channel estimation unit 14, a deframe unit 15, a frequency deinterleave unit 16, and a time deinterleave unit. 17, a demapping unit 18, a Viterbi decoding unit (inner code decoding unit) 19, a byte deinterleaving unit 20, an energy despreading unit 21, and an RS decoding unit (outer code decoding unit) 22. The receiving apparatus 101 receives the OFDM modulated wave of the concatenated code data transmitted from the transmitting apparatus 100 via a receiving antenna, performs processing such as Viterbi decoding and RS decoding, and converts the video / audio data of the original terrestrial digital broadcast into Decrypt.

  The received signal of the OFDM modulated wave received via the receiving antenna is subjected to symbol synchronization processing by the symbol synchronization unit 11, GI is removed by the GI removal unit 12, and the frequency from the time domain signal by the FFT unit 13. Fast Fourier transform is performed on the signal in the region. Then, the propagation path estimation unit 14 estimates the propagation path, and the received signal is equalized to the equalized signal. The equalized signal generated by the propagation path estimator 14 is deframed by the deframer 15, frequency deinterleaved by the frequency deinterleaver 16, and time deinterleaved by the time deinterleaver 17. Then, the demapping unit 18 performs demapping, and a soft decision value is calculated.

  The soft decision value calculated by the demapping unit 18 is Viterbi decoded by the Viterbi decoding unit 19, byte deinterleaved by the byte deinterleaving unit 20, and energy despreading processing is performed by the energy despreading unit 21. The RS decoding unit 22 performs RS decoding. In this manner, terrestrial digital broadcast video / audio data is decoded.

  By the way, in the transmission system configured by the transmission apparatus 100 shown in FIG. 14 and the reception apparatus 101 shown in FIG. 15, in order to improve reception quality, decoding is performed repeatedly on the concatenated code data of the RS code and the convolutional code. Methods are known (see, for example, Non-Patent Documents 1 and 2). For example, the iterative decoding method for the concatenated code data of Non-Patent Document 1 is performed by performing processing of the Viterbi decoding unit 19, the byte deinterleaving unit 20, the energy despreading unit 21, and the RS decoding unit 22 in the receiving apparatus 101 illustrated in FIG. Repeatedly.

  Specifically, the RS decoding unit 22 performs a decoding process on an outer code that is a Reed-Solomon code, performs a syndrome operation by Reed-Solomon decoding, calculates a syndrome value, derives an error position polynomial, and generates an error position and an error A numerical polynomial is obtained, and finally an error numerical value is obtained, error correction is performed, and a decoding result is generated. Then, the RS decoding unit 22 performs the syndrome calculation again to calculate the syndrome value. If the syndrome value is 0, the RS decoding unit 22 generates decoding correct / incorrect information indicating that the correction is successful and the decoding result is correct, and the syndrome value is 0. If not, decoding correct / incorrect information indicating that the correction has failed and the decoding result is not correct is generated.

  The Viterbi decoding unit 19 performs a decoding process on the inner code, which is a convolutional code, and performs decoding by obtaining a state transition path based on data input from the demapping unit 18 using a trellis diagram according to the Viterbi algorithm. Then, the Viterbi decoding unit 19 inputs a decoding result and decoding correctness information from the RS decoding unit 22 when performing an iterative decoding process, and obtains a path of state transition based on these information using a trellis diagram. To decrypt.

Heinea Tsunehiro and two others, “Study on performance improvement by iterative decoding of convolution / erasure Reed-Solomon concatenated codes using soft decision”, IEICE Technical Report, IEICE, November 2008, IEICE Technical Report RCS 2008-149 (2008-11) Shinichi Murata and one other, "Iterative decoding method in error correction decoder of ISDB-T receiver", Proceedings of Communication Lecture 1, 2008 IEICE General Conference, B-5-156, March 18, 2008 Sunday to 21st

  However, in the iterative decoding method of Non-Patent Document 1, the Viterbi decoding unit 19 of the receiving apparatus 101 forcibly sets a path in the trellis diagram based on the decoding result input from the RS decoding unit 22 and the decoding correctness information. Thus, it is considered that decoding is performed. Therefore, in order to realize iterative decoding, it is necessary to modify the normal Viterbi algorithm so that it can be applied to iterative decoding, and the normal Viterbi algorithm cannot be used as it is, and there is a problem that the processing becomes complicated. It was.

  In a general transmission system as well as a transmission system that transmits / receives digital data of terrestrial digital broadcasting, the receiving apparatus repeatedly performs decoding on the concatenated code data without correcting the conventional decoding process, and thereby the reception area. It is desirable that the reception quality can be improved by widening the range and enabling low C / N reception.

  Accordingly, the present invention has been made in view of the above problems, and an object of the present invention is to provide concatenated code data of outer codes and inner codes in a transmission system that transmits and receives digital data such as terrestrial digital television broadcasts and terrestrial digital audio broadcasts. Is to provide a receiving apparatus and a receiving method capable of improving the reception quality with a simple configuration without modifying the conventional decoding process.

In order to achieve the above object, a receiving apparatus according to claim 1 uses a soft decision value of a received signal in a receiving apparatus that repeatedly performs decoding processing on a received signal of concatenated code data including an outer code and an inner code. An inner code decoding unit that performs a decoding process on the inner code, a decoding result obtained by performing an decoding process on the outer code on the signal decoded by the inner code decoding unit, and correcting an error by the decoding process, and The outer code decoding unit that generates decoding correct / incorrect information indicating whether the decoding result is correct, and the decoding correct / incorrect information generated by the outer code decoding unit are subjected to code processing by the inner code, and the inner code decoding correct / incorrect information is obtained. A decoding result coding unit that performs coding processing using the inner code on the decoding result generated by the correct / incorrect information encoding unit to be generated and the outer code decoding unit, and generates an inner code decoding result; Based on the inner code decoding correct / incorrect information generated by the encoding unit, the inner code decoding result generated by the decoding result encoding unit is set to a new soft decision value, and the softness of the received signal used by the inner code decoding unit is set. A soft decision value changing unit that changes a decision value, and the correct / incorrect information encoding unit includes a delay unit that delays the decoding correct / incorrect information generated by the outer code decoding unit, and the decoded correct / incorrect information and the delay unit. An adder for adding the delayed decoding correct / incorrect information, the addition result by the adder is generated as inner code decoding correct / incorrect information, and the soft decision value changing unit is generated by the correct / incorrect information encoding unit. When the decoding correct / incorrect information is a predetermined value, when the inner code decoding result generated by the decoding result encoding unit is 1, a new soft decision value is set to 1, and when the inner code decoding result is 0, a new 0 for soft decision value Set, changes the soft decision value of the received signal used in the inner code decoding section, it is characterized.

The receiving apparatus according to claim 2, on the received signal of concatenated codes data including outer code and inner code, a receiving apparatus to repeat the decoding process to extract the pilot signal from the received signal, a predetermined amplitude and phase A propagation path of the extracted pilot signal is calculated using a pilot signal having the following, and a propagation path of the reception signal is estimated by an interpolation process using the propagation path of the pilot signal, and the propagation path of the reception signal is used. A propagation path estimator that equalizes the received signal and generates an equalized signal, and a demultiplexer that demaps the I and Q signals of the equalized signal generated by the propagation path estimator and generates a soft decision value. A mapping unit, an inner code decoding unit that performs a decoding process on the inner code using the soft decision value of the equalized signal generated by the demapping unit, and a decoding process performed by the inner code decoding unit. An outer code decoding unit that performs a decoding process on the outer code on the received signal, generates a decoding result that has been error-corrected by the decoding process, and decoding correct / incorrect information that indicates whether the decoding result is correct, and the outer code decoding The decoding correct / incorrect information generated by the unit is subjected to the encoding process by the inner code, and the correct / incorrect information encoding unit for generating the inner code decoding correct / incorrect information and the decoding result generated by the outer code decoding unit by the inner code A decoding result encoding unit that performs an encoding process and generates an inner code decoding result, inner code decoding correct / incorrect information generated by the correct / incorrect information encoding unit, and an inner code decoding result generated by the decoding result encoding unit, A mapping unit that maps on the Q axis, and the propagation path estimation unit has a correct decoding result in the inner code decoding correct / incorrect information mapped by the mapping unit Mapping on the I and Q axes including the inner code decoding result corresponding to the identified inner code decoding correct / incorrect information among the inner code decoding results mapped by the mapping unit A point candidate is selected, and based on a predetermined received signal corresponding to the inner code decoding correct / incorrect information indicating that the mapping point candidate and the decoding result are correct, the predetermined one of the mapping point candidates A mapping point closest to the reception signal is selected, a propagation path of the predetermined reception signal is calculated from the selected mapping point and the predetermined reception signal, and a propagation path of the pilot signal and a propagation path of the predetermined reception signal The received signal propagation path is estimated by interpolation using the received signal, the received signal propagation path is equalized, and the received signal is equalized. A new post-equalization signal used in the mapping unit is generated.

The receiving apparatus according to claim 3 is the receiving apparatus according to claim 2 , wherein the correct / incorrect information encoding unit delays the decoding correct / incorrect information generated by the outer code decoding unit, the decoded correct / incorrect information, and the An adder for adding the decoding correct / incorrect information delayed by the delay unit, and generating an addition result by the adder as inner code decoding correct / incorrect information, wherein the propagation path estimation unit is mapped by the mapping unit The code decoding correct / incorrect information is an information indicating whether the decoding result is correct, and an inner code decoding correct / incorrect information checking unit that checks whether the decoding result is correct. A mapping point for selecting a mapping point candidate arranged on the I and Q axes, including an inner code decoding result corresponding to the inner code decoding correct / incorrect information. A mapping point candidate selected by the mapping point candidate selection unit, the mapping point candidate selected by the mapping point candidate selection unit, and a predetermined received signal corresponding to the inner code decoding correct / incorrect information indicating that the decoding result is correct A mapping point selection unit that selects a mapping point closest to the predetermined reception signal among the candidates, and complex division of the predetermined reception signal by the mapping point selected by the mapping point selection unit, A first division unit for calculating a propagation path of the received signal, an extraction unit for extracting a pilot signal from the received signal, and a pilot signal extracted by the extraction unit as a pilot signal having a predetermined amplitude and phase A second division unit for performing complex division and calculating a propagation path of the pilot signal; and the first division. An interpolation unit that estimates a new propagation path of the reception signal by linear interpolation processing using the propagation path of the predetermined reception signal calculated by the unit and the propagation path of the pilot signal calculated by the second division unit; And a third dividing unit that complex-divides the received signal by a propagation path of the received signal estimated by the interpolating unit to generate a new equalized signal.

The receiving device according to claim 4 is the receiving device according to claim 2 , and further, performs decoding processing on the decoding correct / incorrect information generated by the outer code decoding unit by the inner code to generate inner code decoding correct / incorrect information. And a second decoding result encoding unit that performs an encoding process using the inner code on the decoding result generated by the outer code decoding unit, and generates an inner code decoding result; The inner code decoding result generated by the second decoding result encoding unit is set to a new soft decision value based on the inner code decoding correctness information generated by the two correct / incorrect information encoding units, and the inner code decoding unit And a soft decision value changing unit for changing the soft decision value used in.

According to a fifth aspect of the present invention, there is provided a first method of calculating a soft decision value of the received signal in a receiving method in which decoding processing is repeatedly performed on a received signal of concatenated code data including an outer code and an inner code. And a second step of performing a decoding process on the inner code using the soft decision value, and a decoding result obtained by performing an decoding process on the outer code on the signal decoded in the second step and correcting an error. And a third step for generating decoding correct / incorrect information indicating whether or not the decoding result is correct, and a decoding process for determining whether the decoding result is correct or not by performing a code process using the inner code on the decoding correct / incorrect information generated in the third step. A fourth step of generating information; and a fifth step of performing an inner code decoding result on the decoding result generated by the third step to generate an inner code decoding result; Based on the inner code decoding correct / incorrect information generated in the fourth step, the inner code decoding result generated in the fifth step is set as a new soft decision value and used in the second step. And the fourth step delays the decoding correct / incorrect information generated in the third step, and adds the decoded correct / incorrect information and the delayed decoded correct / incorrect information. The addition result is generated as inner code decoding correct / incorrect information, and the sixth step generates the inner code decoding correct / incorrect information generated in the fifth step when the inner code decoding correct / incorrect information generated in the fourth step is a predetermined value. When the code decoding result is 1, a new soft decision value is set to 1. When the inner code decoding result is 0, a new soft decision value is set to 0, and the received signal softness in the first step is set. Judgment value Further to, characterized in that.

According to a sixth aspect of the present invention, there is provided a receiving method for extracting a pilot signal from the received signal by repeating decoding processing on a received signal of concatenated code data including an outer code and an inner code. A propagation path of the extracted pilot signal is calculated using a pilot signal having a phase, a propagation path of the reception signal is estimated by interpolation processing using the propagation path of the pilot signal, and a propagation path of the reception signal is used. A first step of equalizing the received signal and generating an equalized signal; and a second step of demapping the equalized signals I and Q generated in the first step to generate a soft decision value Step, a third step of performing a decoding process on the inner code using the soft decision value generated in the second step, and a signal decoded by the third step A fourth step of performing a decoding process on the outer code, generating a decoding result error-corrected by the decoding process, and decoding correct / incorrect information indicating whether or not the decoding result is correct, and generated by the fourth step The decoding process is performed by the inner code on the decoded correct / incorrect information, the inner code decoding correct / incorrect information is generated in the fifth step, and the decoding process generated by the fourth step is performed by the inner code. A sixth step of generating an inner code decoding result, an inner code decoding correct / incorrect information generated by the fifth step, and an inner code decoding result generated by the sixth step are mapped on the I and Q axes. Among the inner code decoding correct / incorrect information mapped in step 7 and the seventh step, the inner code decoding correct / incorrect information indicating that the decoding result is correct is specified. 8 including the inner code decoding result corresponding to the inner code decoding correct / incorrect information identified in the eighth step out of the inner code decoding results mapped by the seventh step. Based on the ninth step of selecting a mapping point candidate, the mapping point candidate selected in the ninth step, and the predetermined received signal corresponding to the inner code decoding correctness information specified in the eighth step A tenth step of selecting a mapping point closest to the predetermined received signal among mapping point candidates selected in the ninth step; a mapping point selected in the tenth step; and the predetermined point An eleventh step of calculating a propagation path of the predetermined received signal from the received signal, and the eleventh step. A new post-equalization signal used in the second step by estimating a new propagation path of the reception signal by interpolation processing using the propagation path of the predetermined reception signal calculated by the step and the propagation path of the pilot signal And generating a twelfth step.

  As described above, according to the present invention, when iterative decoding is performed on concatenated code data of an outer code and an inner code, a soft decision is made based on the decoding result and decoding correct / incorrect information indicating whether the decoding result is correct. The value or / and the propagation path estimation value are changed. As a result, the quality of the received signal can be improved with a simple configuration without modifying the conventional decoding process.

It is a block diagram which shows the structure of the receiver by the 1st Embodiment (Example 1) of this invention. It is a block diagram which shows the structure of the convolutional code | cord | chord part which convolutionally codes a decoding result. It is a block diagram which shows the structure of the convolutional code | cord | chord part which convolutionally encodes decoding correctness information. It is an input / output diagram explaining the process of a soft decision value change part. It is a flowchart explaining the process of a soft decision value change part. It is a specific example explaining the process of a soft decision value change part. It is a figure which shows the simulation result of the static characteristic by Example 1. FIG. It is a figure which shows the simulation result of the dynamic characteristic by Example 1. FIG. It is a block diagram which shows the structure of the receiver by the 2nd Embodiment (Example 2) of this invention. It is a block diagram which shows the structure of a propagation path estimation part. It is a flowchart explaining the process of a propagation path estimation part. It is a figure explaining the process of a propagation path estimation part. It is a block diagram which shows the structure of the receiver by the 3rd Embodiment (Example 3) of this invention. It is a block diagram which shows the structure of a transmitter. It is a block diagram which shows the structure of the conventional receiver. It is a block diagram which shows the structure of the conventional propagation path estimation part. It is a flowchart explaining the process of the conventional propagation path estimation part. It is a figure explaining the process of the conventional propagation path estimation part.

  The best mode for carrying out the present invention will be described below with reference to the drawings. When the reception apparatus according to the first embodiment (Example 1) of the present invention performs iterative decoding using concatenated code data of an RS code and a convolutional code using a soft decision value, a decoding result by RS decoding, and an RS The quality of the received signal is improved by changing the soft decision value based on the decoding correct / incorrect information indicating whether or not the decoding result is correct. In addition, when the receiving apparatus according to the second embodiment (Example 2) of the present invention performs iterative decoding on concatenated code data of an RS code and a convolutional code, a decoding result by RS decoding and a decoding result by RS decoding The quality of the received signal is improved by changing the propagation path estimation value based on decoding correct / incorrect information indicating whether or not is correct. Further, the receiving apparatus according to the third embodiment (Example 3) of the present invention improves the quality of the received signal by combining the Example 1 and Example 2 and changing the soft decision value and the propagation path estimation value. It is something to be made. Hereinafter, Examples 1 to 3 will be described.

  First, the receiving apparatus according to the first embodiment will be described. In the first embodiment, when iterative decoding is performed on the concatenated code data of the RS code and the convolutional code using the soft decision value, the decoding result indicating whether the decoding result by the RS decoding and the decoding result of the RS decoding are correct or not is shown. The soft decision value is changed based on the information.

  FIG. 1 is a block diagram illustrating a configuration of a receiving apparatus according to the first embodiment. This receiver 1 includes a symbol synchronization unit 11, a GI removal unit 12, an FFT unit 13, a propagation path estimation unit 14, a deframe unit 15, a frequency deinterleave unit 16, a time deinterleave unit 17, a demapping unit 18, and a repetition A decoding unit 41 is provided.

  When receiving apparatus 1 receives an OFDM modulated wave of concatenated code data based on an RS code and a convolutional code from transmitting apparatus 100 shown in FIG. 14, symbol synchronization section 11 receives the received signal of the OFDM modulated wave via the receiving antenna. The received signal is subjected to symbol synchronization processing, and the received signal for each symbol is output to the GI removing unit 12.

  The GI removal unit 12 receives the reception signal for each symbol from the symbol synchronization unit 11, removes the GI from the reception signal corresponding to the GI addition unit 80, and outputs the GI to the FFT unit 13. The FFT unit 13 receives the received signal from which the GI has been removed from the GI removing unit 12, and performs fast Fourier transform on the received signal in the time domain into a received signal in the frequency domain corresponding to the IFFT unit 79, and a propagation path estimating unit 14 for output.

  The propagation path estimation unit 14 receives a frequency domain received signal from the FFT unit 13, extracts an SP (Scattered Pilot) from the received signal, and transmits the SP (transmitted device 100) set in advance. SP propagation path is calculated based on (SP) and the extracted SP, and the propagation path of the received signal is estimated by interpolation processing. Then, the propagation path estimation unit 14 calculates an equalized signal based on the estimated propagation path of the received signal and the received signal, and outputs the calculated signal to the deframe unit 15. Details of the propagation path estimation unit 14 will be described later.

  The deframe unit 15 receives the equalized signal from the propagation path estimation unit 14, performs deframe processing corresponding to the OFDM frame unit 78, and outputs the deframe to the frequency deinterleave unit 16. The frequency deinterleaver 16 receives the equalized signal (received signal) deframed from the deframer 15, performs frequency deinterleave corresponding to the frequency interleaver 77, and outputs it to the time deinterleaver 17. To do. The time deinterleave unit 17 receives the frequency deinterleaved reception signal from the frequency deinterleave unit 16, performs time deinterleave corresponding to the time interleave unit 76, and outputs the result to the demapping unit 18.

  The demapping unit 18 receives the time-deinterleaved received signal from the time deinterleaving unit 17, performs demapping corresponding to the mapping unit 75, and calculates a soft decision value (a value in the range from 0 to 1). And output to the iterative decoding unit 41. Specifically, the demapping unit 18 calculates a soft decision value by soft decision processing, for example, by calculating a distance between an input received signal on the I and Q axes and a preset mapping point. . Here, the soft decision value takes a value in a range from 0 to 1, and the closer to 0, the higher the probability that the signal transmitted by the transmission apparatus 100 is 0, and the closer to 1, the higher the probability of being 1. The closer to 0.5, the higher the probability that it is unknown whether it is 0 or 1.

[Iterative decoding unit]
The iterative decoding unit 41 includes a soft decision value changing unit 50, a Viterbi decoding unit 19, a byte deinterleaving unit 20, an energy despreading unit 21, an RS decoding unit 22, byte interleaving units 31 and 34, a convolutional code unit (correct / uncorrection information code unit). ) 32, a convolutional code unit (decoding result code unit) 35, and an energy spreading unit 33. The iterative decoding unit 41 receives the soft decision value from the demapping unit 18, changes the soft decision value based on the decoding result and decoding correctness information of the video / audio data generated by the decoding process, and uses the changed soft decision value Then, the decoding process is performed, and the change of the soft decision value and the decoding process are repeated a predetermined number of times to generate video / audio data with few errors. The soft decision value changing unit 50, the Viterbi decoding unit 19, and the convolutional coding units 32 and 35 perform processing in units of bits, and include a byte deinterleaving unit 20, an energy despreading unit 21, an RS decoding unit 22, a byte interleaving unit 31, 34 and the energy diffusion unit 33 perform processing in units of bytes.

  The soft decision value changing unit 50 receives the soft decision value from the demapping unit 18, inputs convolution decoding correct / incorrect information from the convolutional code unit 32, and convolution decoding result from the convolutional code unit 35. The soft decision value is changed based on the convolution decoding result, and the changed soft decision value is output to the Viterbi decoding unit 19. When the first iterative decoding process (first decoding process) is performed by the iterative decoding unit 41, the soft decision value changing unit 50 outputs the input soft decision value as it is. On the other hand, the soft decision value changing unit 50 outputs the changed soft decision value when the iterative decoding unit 41 performs iterative decoding processing from the second time to a predetermined number of times. Details of the soft decision value changing unit 50 will be described later.

  The Viterbi decoding unit 19 receives the soft decision value from the soft decision value changing unit 50, performs Viterbi decoding as decoding processing corresponding to the inner code in the convolutional coding unit 74, and outputs it to the byte deinterleaving unit 20. The byte deinterleave unit 20 receives the Viterbi decoding result obtained by Viterbi decoding from the Viterbi decoding unit 19, performs byte deinterleaving corresponding to the byte interleaving unit 73, and outputs the result to the energy despreading unit 21. The energy despreading unit 21 receives the Viterbi decoding result subjected to byte deinterleaving from the byte deinterleaving unit 20, performs energy despreading processing corresponding to the energy spreading unit 72, and outputs the result to the RS decoding unit 22.

  The RS decoding unit 22 receives the Viterbi decoding result subjected to the energy despreading process from the energy despreading unit 21, performs RS decoding as a decoding process corresponding to the outer code in the RS encoding unit 71, and outputs the decoding result of the video / audio data. The data is output to the energy diffusing unit 33, and the decoding correctness information is output to the byte interleaving unit 31. In addition, when the iterative decoding unit 41 completes a predetermined number of iterative decoding processes, the RS decoding unit 22 outputs the decoding result obtained by RS decoding to the outside as video / audio data. On the other hand, when the iterative decoding unit 41 has not completed the predetermined number of times of iterative decoding processing, the RS decoding unit 22 sends the decoding result of the RS-decoded video / audio data to the energy spreading unit 33 and the decoding correctness information to the byte interleave. Output to the unit 31.

  Specifically, the RS decoding unit 22 performs a syndrome operation by RS decoding on the input Viterbi decoding result after byte deinterleaving, calculates a syndrome value, derives an error position polynomial, and generates an error position and an error numerical value. A polynomial is obtained, and finally, an error value is obtained and error correction is performed, and a decoding result of the video / audio data is generated. The RS decoding unit 22 outputs the decoding result of the video / audio data to the energy diffusing unit 33 when the predetermined number of times of the iterative decoding process has not been completed. Output to. In addition, when the predetermined number of repeated decoding processes have not been completed, the RS decoding unit 22 performs the syndrome calculation again to calculate the syndrome value. When the syndrome value is 0, the correction is successful and the decoding result is correct. Decoding correct / incorrect information “1” is generated, and if the syndrome value is not 0, the correct / uncorrected information “0” indicating that the correction has failed and the decoding result is incorrect is generated. 1) is output to the byte interleave unit 31.

  The energy spreading unit 33 receives the decoding result of the video / audio data from the RS decoding unit 22, performs the same energy spreading process as the energy spreading unit 72, and outputs the result to the byte interleaving unit 34. The byte interleaving unit 34 receives the decoding result subjected to the energy spreading process from the energy spreading unit 33, performs the same byte interleaving as the byte interleaving unit 73, and outputs the result to the convolutional coding unit 35.

  The convolutional code unit 35 receives the decoding result that has been byte-interleaved from the byte interleaving unit 34, performs code processing using an inner code that is the same convolutional code as the convolutional code unit 74, and performs a soft decision on the convolutional decoding result (0, 1). The value is output to the value changing unit 50.

  FIG. 2 is a block diagram showing a configuration of the convolutional coding unit 35 shown in FIG. This convolutional coding unit 35 uses a coder with a constraint length of 7 and includes six delay units (D) and two XOR units, and includes a convolutional coding unit 74 of the transmission apparatus 100 shown in FIG. It has the same configuration. The six delay units delay the input decoding result (0, 1) by 1 bit each. The first XOR unit receives the decoding result from the byte interleaving unit 34 and the decoding result output from each of the first, second, third, and sixth delay units, performs an exclusive OR operation, and performs convolutional decoding. The result X (0, 1) is output to the soft decision value changing unit 50. The second XOR unit inputs the decoding result from the byte interleaving unit 34 and the decoding result output from the second, third, fifth, and sixth delay units, respectively, and performs an exclusive OR operation. The convolution decoding result Y (0, 1) is output to the soft decision value changing unit 50.

  Returning to FIG. 1, the byte interleaving unit 31 inputs the decoding correct / incorrect information (0, 1) from the RS decoding unit 22, and the input decoding correct / incorrect information (0) along with the byte interleaving for the decoding result in the byte interleaving unit 34. , 1) is subjected to the same byte interleaving and output to the convolutional code unit 32.

  The convolutional coding unit 32 receives the decoding correct / incorrect information (0, 1) byte-interleaved from the byte interleaving unit 31, and the convolutional coding for the decoding result in the convolutional coding unit 35 causes the input decoding correct / incorrect information (0, 1). 1), a code process using an inner code that is a convolutional code is performed, and convolutional decoding correctness information (0 to 5) is output to the soft decision value changing unit 50.

  FIG. 3 is a block diagram showing the configuration of the convolutional coding unit 32 shown in FIG. The convolutional code unit 32 includes six delay units (D) and two adders. Comparing the convolutional coding unit 74 of the transmission apparatus 100 shown in FIG. 14 and the convolutional coding unit 35 shown in FIG. 2 with the convolutional coding unit 32, they are the same in that an encoder with a constraint length of 7 is used. However, the convolutional code unit 32 is different in that an adder is provided instead of the XOR unit in the convolutional code units 35 and 74. The six delay units delay the inputted decoding correctness information (0, 1) by 1 bit each. The first adder receives the decoding correct / incorrect information from the byte interleave unit 31 and the decoding correct / incorrect information output from the first, second, third, and sixth delay units, performs an addition operation, and performs convolution decoding correct / incorrect information. X (0 to 5) is output to the soft decision value changing unit 50. Here, the convolutional decoding correctness information X is a value of 0, 1, 2, 3, 4, or 5 depending on the value of the input decoding correctness information. The second adder inputs the decoding correct / incorrect information from the byte interleave unit 31 and the decoding correct / incorrect information output from the second, third, fifth, and sixth delay units, performs an addition operation, and performs convolutional decoding. The correctness information Y (0 to 5) is output to the soft decision value changing unit 50. The convolutional decoding correct / incorrect information Y, like the convolutional decoding correct / incorrect information X, has a value of 0, 1, 2, 3, 4, or 5.

  Returning to FIG. 1, when the first iterative decoding process (first decoding process) is performed as described above, the soft decision value changing unit 50 directly performs the Viterbi decoding on the soft decision value input from the demapping unit 18. To the unit 19. Further, as described above, the soft decision value changing unit 50, when the iterative decoding process from the second time to the predetermined number of times is performed, the convolution decoding result X, Y (0, 1) input from the convolution coding unit 35, Based on the convolution decoding correct / incorrect information X, Y (0 to 5) input from the convolutional coding unit 32, the input soft decision value is changed to a new soft decision value, and the new soft decision value is sent to the Viterbi decoding unit 19. Output. Hereinafter, the convolution decoding results X and Y, which are the results of puncturing at a coding rate corresponding to the convolutional coding unit 74, are simply referred to as convolutional decoding results, and are the results of puncturing at the coding rate corresponding to the convolutional coding unit 74. The convolution decoding correct / incorrect information X and Y is simply referred to as convolution decoding correct / incorrect information. When the convolution decoding correct / incorrect information is 5, the soft decision value changing unit 50 sets a new soft decision value to 0 when the corresponding convolution decoding result is 0, and sets a new soft decision value when the convolution decoding result is 1. When the decision value is set to 1 and output, and the convolutional decoding correctness information is not 5, the inputted soft decision value is output as it is without being changed.

  The process of the soft decision value changing unit 50 when the iterative decoding process from the second time to the predetermined number of times is performed will be described in detail. 4 is an input / output diagram for explaining the processing of the soft decision value changing unit 50, FIG. 5 is a flowchart for explaining the processing of the soft decision value changing unit 50, and FIG. It is a specific example explaining the process.

  Referring to FIG. 4, soft decision value changing unit 50 receives soft decision value a from demapping unit 18, convolution decoding correct / incorrect information b from convolution coding unit 32, and convolution decoding result c from convolution coding unit 35. To do. Then, the soft decision value changing unit 50 executes the program shown in the location of the soft decision value changing unit 50 in FIG. 4 to perform the process of the flowchart shown in FIG. 5 to convert the soft decision value a into the soft decision value d. And the soft decision value d is output to the Viterbi decoding unit 19.

  Referring to FIG. 5, soft decision value changing unit 50 determines whether or not convolution decoding correct / incorrect information b is 5 (step S501). When it is determined that the convolutional decoding correctness information b is 5 (step S501: Y), the process proceeds to step S502. When it is determined that the convolutional decoding correctness information b is not 5 (step S501: N), from the demapping unit 18 The input soft decision value a is set as it is as the soft decision value d (step S503), and the process proceeds to step S506.

  When the third and subsequent iterations are performed, the soft decision value a is equal to the soft decision value d (soft decision value a input from the demapping unit 18) in step S503 or the later-described decoding processing. The soft decision value d changed in step S504 or step S506 is used.

  The soft decision value changing unit 50 proceeds from step S501 to determine whether the convolution decoding result c is 1 or 0 (step S502). When it is determined that the convolutional decoding result c is 1 (step S502: = 1), the soft decision value is changed by substituting 1 as the convolutional decoding result c into the soft decision value d regardless of the soft decision value a. (Step S504), the process proceeds to Step S506. On the other hand, when it is determined that the convolutional decoding result c is 0 (step S502: = 0), 0 which is the convolutional decoding result c is substituted for the soft decision value d regardless of the soft decision value a, and the soft decision value is set. Change (step S505), the process proceeds to step S506.

  The soft decision value changing unit 50 proceeds from step S503 to step S505, and outputs the soft decision value d to the Viterbi decoding unit 19 (step S506).

  Referring to FIG. 6, for example, when convolution decoding correct / incorrect information b is 0, 2, 3, or 4, soft decision value changing unit 50 determines that soft decision value change processing is not performed (change processing determination). X) Regardless of the value of the convolution decoding result c, the input soft decision value a is set as the soft decision value d and output. On the other hand, when the convolutional decoding correct / incorrect information b is 5, the soft decision value changing unit 50 determines that the soft decision value changing process is to be performed (change processing determination ○), and the convolution decoding result regardless of the soft decision value a. The soft decision value is changed by substituting the value of c into the soft decision value d, and the soft decision value d is output.

〔simulation result〕
Next, the simulation result by the receiving apparatus 1 of Example 1 shown in FIG. 1 is demonstrated. FIG. 7 is a diagram illustrating simulation results of static characteristics (Gaussian noise characteristics) according to the first embodiment. FIG. 8 illustrates dynamic characteristics (characteristics in the case of a six-wave Rayleigh model and a maximum Doppler frequency of 10 Hz) according to the first embodiment. It is a figure which shows a simulation result. The simulation results of FIG. 7 and FIG. 8 show the case where the carrier modulation is 64QAM, the coding rate is R = 3/4, the time interleave I = 2, and the number of repetition decoding ite = 0, 1, 5. 7 and 8, the horizontal axis represents C / N (Carrier Noise Ratio), and the vertical axis represents BER (Bit Error Rate).

  According to the simulation result of the static characteristics according to Example 1 of FIG. 7, when BER = 2.0e-4, the C / N when the iterative decoding number it = 1, 5 is the iterative decoding number it = 0 (repetition). It can be seen that there is an improvement of about 1.3 dB compared to the one-time decoding process in which no decoding process is performed.

  According to the simulation result of the dynamic characteristic according to the first embodiment of FIG. 8, when BER = 2.0e-4, the C / N when the iterative decoding number it = 1 is larger than the iterative decoding number it = 0. It can be seen that there is an improvement of about 4.4 dB. In addition, it can be seen that the C / N when the number of iterations iteration = ite is improved by about 4.7 dB compared to the number of iterations iteration_it = 0.

  As described above, according to the receiving device 1 of the first embodiment illustrated in FIG. 1, the iterative decoding unit 41 performs iterative decoding using concatenated code data of RS code and convolutional code using soft decision values. The soft decision value is changed based on the decoding result by RS decoding and the decoding correct / incorrect information indicating whether the decoding result of RS decoding is correct. Specifically, the RS decoding unit 22 generates a decoding result and decoding correctness information, and the convolutional code unit 35 performs an inner code process using a convolutional code on the energy-spread and byte-interleaved decoding result, thereby performing convolutional decoding. The result is generated, and the convolutional coding unit 32 performs the inner code processing using the convolutional code on the byte-interleaved decoding correctness information to generate the convolutional decoding correctness information in the range of 0 to 5. Then, the soft decision value changing unit 50 changes the convolution decoding result corresponding to the convolution decoding correct / incorrect information 5 to a new soft decision value instead of the soft decision value input from the demapping unit 18. The iterative decoding unit 41 performs a decoding process using the new soft decision value changed in this way, generates video / audio data by a predetermined number of decoding processes, and outputs it to the outside.

  Here, the convolutional coding unit 32 generates convolutional decoding correct / incorrect information from one decoding correct / incorrect information within a range that affects the front and rear on the time axis. When the predetermined decoding correct / incorrect information is 1 within a range that affects the front and rear on the time axis, a value of 5 is generated as the convolution decoding correct / incorrect information. That is, when the convolutional decoding correctness information is a value of 5, the corresponding decoding result is a correct result without being affected by other decoding results before and after on the time axis, and an incorrect result. Is not included. Since a new soft decision value is generated based on such a decoding result, and the decoding process is performed using the new soft decision value, the iterative decoding unit 41 can generate video / audio data with few errors. .

  The iterative decoding unit 41 includes the Viterbi decoding unit 19, the byte deinterleaving unit 20, the energy despreading unit 21, and the RS decoding unit 22 that constitute the conventional decoding processing unit as they are, and a new soft decision value changing unit 50 and the like. I was prepared to. Thereby, it is possible to perform iterative decoding without correcting the decoding process such as the conventional Viterbi algorithm. Therefore, when iterative decoding is performed on the concatenated code data of the RS code and the convolutional code, the quality of the received signal can be improved with a simple configuration without modifying the conventional decoding process. That is, the reception area can be expanded and low C / N reception is possible.

  Next, a receiving apparatus according to the second embodiment will be described. In the second embodiment, when iteratively decoding the concatenated code data of the RS code and the convolutional code, the propagation path based on the decoding result by RS decoding and the decoding correct / incorrect information indicating whether the decoding result of RS decoding is correct or not The estimated value is changed.

  FIG. 9 is a block diagram illustrating a configuration of a receiving apparatus according to the second embodiment. The receiving device 2 includes a symbol synchronization unit 11, a GI removal unit 12, an FFT unit 13, and an iterative decoding unit 42. When the receiving apparatus 1 of the first embodiment shown in FIG. 1 and the receiving apparatus 2 of the second embodiment shown in FIG. 9 are compared, the symbol synchronizer 11, the GI removing unit 12, and the FFT unit 13 are both included in the receiving devices 1 and 2. The receiving apparatus 1 includes the iterative decoding unit 41 that performs iterative decoding by changing the soft decision value, while the receiving apparatus 2 re-estimates the propagation path. The difference is that an iterative decoding unit 42 that performs iterative decoding using a new equalized signal is provided. Referring to FIG. 9, symbol synchronization unit 11, GI removal unit 12, and FFT unit 13 are the same as those shown in FIG.

[Iterative decoding unit]
The iterative decoding unit 42 includes a channel estimation unit 60, a deframe unit 15, a frequency deinterleave unit 16, a time deinterleave unit 17, a demapping unit 18, a Viterbi decoding unit 19, a byte deinterleave unit 20, and an energy despreading unit. 21, RS decoding unit 22, byte interleaving units 31 and 34, convolutional coding units 32 and 35, energy spreading unit 33, remapping unit 36, time interleaving unit 37, frequency interleaving unit 38 and OFDM framing unit 39. . The iterative decoding unit 42 receives the received signal from the FFT unit 13, re-estimates the propagation path based on the decoding result and decoding correctness information of the video / audio data generated by the decoding process, and calculates a new equalized signal. , Decoding using the new equalized signal, and re-estimating the propagation path, calculating the new equalized signal and repeating the decoding process a predetermined number of times to generate video and audio data with few errors To do. Note that the propagation path estimation unit 60, the deframing unit 15, the frequency deinterleaving unit 16, the time deinterleaving unit 17, the demapping unit 18, the Viterbi decoding unit 19, the convolutional coding units 32 and 35, the remapping unit 36, and the time interleaving. Unit 37, frequency interleaving unit 38, and OFDM framing unit 39 perform processing in units of bits, and byte deinterleaving unit 20, energy despreading unit 21, RS decoding unit 22, byte interleaving units 31, 34, and energy spreading unit 33 are Perform byte-by-byte processing.

  The propagation path estimation unit 60 receives the reception signal from the FFT unit 13 and also receives the convolutional decoding result and convolutional decoding correct / incorrect information that are remapping signals from the OFDM framing unit 39, and receives the reception signal, the convolutional decoding result, and the convolutional decoding. Based on the correctness information, the propagation path is estimated again, a new equalized signal is calculated and output to the deframer 15. When the iterative decoding unit 42 performs the first iterative decoding process (first decoding process), the propagation path estimation unit 60 estimates the propagation path from the input received signal and calculates an equalized signal. Output. On the other hand, when the iterative decoding unit 42 performs the iterative decoding process from the second time to the predetermined number of times, the propagation path estimation unit 60 estimates the propagation path again, calculates a new equalized signal, A new equalized signal is output. Details of the propagation path estimation unit 60 will be described later.

  Deframing unit 15, frequency deinterleaving unit 16, time deinterleaving unit 17, demapping unit 18, Viterbi decoding unit 19, byte deinterleaving unit 20, energy despreading unit 21, RS decoding unit 22, byte interleaving unit 31, 34, the convolutional code units 32 and 35, and the energy spreading unit 33 are the same as those shown in FIG.

  The remapping unit 36 receives the convolution decoding correct / incorrect information (0 to 5) from the convolutional coding unit 32 and the convolutional decoding result (0, 1) from the convolutional coding unit 35, and the convolutional decoding result (0, 1). On the other hand, the same mapping as that of the mapping unit 75 is performed (refer to a QPSK (Quadrature Phase Shift Keying) mapping example in FIG. 12 to be described later) and output to the time interleaving unit 37. The processing by the remapping unit 36 is remapping in that the mapping unit 75 of the transmission device 100 illustrated in FIG. 14 performs mapping and the demapping unit 18 of the reception device 2 performs demapping. Further, the remapping unit 36 associates the input convolution decoding correct / incorrect information (0 to 5) with the mapping point of the convolution decoding result in accordance with the mapping with respect to the convolution decoding result (0, 1), and performs time interleaving. To the unit 37.

  The time interleaving unit 37 receives the convolutional decoding result (0, 1) and the convolutional decoding correct / incorrect information (0-5) remapped from the remapping unit 36, and performs time interleaving on the convolutional decoding result (0, 1). The same time interleaving as that of the unit 76 is performed and output to the frequency interleaving unit 38. Further, the time interleaving unit 37 performs the same time interleaving on the input convolutional decoding correctness information (0 to 5) with the time interleaving for the convolutional decoding result, and outputs the result to the frequency interleaving unit 38.

  The frequency interleaving unit 38 receives the convolutional decoding result (0, 1) and the convolutional decoding correct / incorrect information (0-5) that are time interleaved from the time interleaving unit 37, and performs frequency interleaving on the convolutional decoding result (0, 1). The same frequency interleaving as that of the unit 77 is performed and output to the OFDM framing unit 39. Further, the frequency interleaving unit 38 performs the same frequency interleaving on the input convolutional decoding correct / incorrect information (0 to 5) in accordance with the frequency interleaving for the convolutional decoding result (0, 1), and sends it to the OFDM framing unit 39. Output.

  The OFDM framing unit 39 inputs the convolutional decoding result (0, 1) and the convolutional decoding correct / incorrect information (0 to 5) frequency-interleaved from the frequency interleaving unit 38, and the OFDM framing unit 39 applies OFDM to the convolutional decoding result (0, 1). The same OFDM framing as the framing unit 78 is performed and output to the propagation path estimation unit 60. Further, the OFDM framing unit 39 performs the same OFDM framing on the input convolutional decoding correct / incorrect information (0 to 5) with the OFDM framing for the convolutional decoding result (0, 1) to estimate the propagation path. To the unit 60.

  As described above, when the first iterative decoding process (first decoding process) is performed, the propagation path estimation unit 60 estimates the propagation path from the received signal input from the FFT unit 13 and calculates the equalized signal. And output to the deframing unit 15. Further, the propagation path estimation unit 60 receives the received signal input from the FFT unit 13 and the convolutional decoding result (0, 1) input from the OFDM framing unit 39 when iterative decoding processing from the second time to the predetermined number of times is performed. ) And convolutional decoding correctness information (0 to 5), the propagation path is estimated again, a new equalized signal is calculated and output to the deframer 15. Specifically, when the convolution decoding correct / incorrect information is 5, the propagation path estimation unit 60 uses the mapping point of the convolution decoding result (0, 1) at that time and the mapping point of the received signal, and Calculate the propagation path. Then, the propagation path estimation unit 60 performs interpolation processing using the propagation path in the received signal and the propagation path calculated by the SP, estimates the propagation path of the received signal, and calculates a new equalized signal. Hereinafter, the propagation path estimation unit 60 will be described in detail.

(Conventional propagation path estimation process)
Prior to describing the processing of the propagation path estimation unit 60 in the case where iterative decoding processing is performed from the second time to the predetermined number of times, first, conventional propagation path estimation processing will be described. FIG. 16 is a block diagram showing a configuration of a propagation path estimation unit 14 that performs a conventional propagation path estimation process, FIG. 17 is a flowchart for explaining the process of the propagation path estimation unit 14, and FIG. It is a figure explaining the process of the estimation part. In addition, the propagation path estimation part 60 in Example 2 performs the same process as the propagation path estimation part 14, when the first iterative decoding process (first decoding process) is performed.

  Referring to FIG. 16, the conventional propagation path estimation unit 14 includes an SP generation unit 141, an SP extraction unit 142, division units 143 and 145, and an interpolation unit 144. 16 to 18, the SP generation unit 141 generates a known SP carrier (SP carrier having a predetermined amplitude and phase) transmitted by the transmission apparatus 100, and outputs the generated SP carrier to the division unit 143 as a generated SP. (Step S1701). The SP extraction unit 142 extracts only the SP carrier from the input reception signal, and outputs the extracted reception SP to the division unit 143 (step S1702). The division unit 143 receives the generated SP from the SP generation unit 141 and the reception SP from the SP extraction unit 142, performs complex division on the reception SP by the generation SP, and outputs the divided SP to the interpolation unit 144 (step S1703). In this way, the SP propagation path value is calculated as the post-division SP. The SP carriers are arranged at intervals of 12 carriers in one symbol. For this reason, the propagation path estimated value of the received signal is estimated based on the SP propagation path value.

  The interpolation unit 144 receives the post-division SP (SP propagation path value) from the division unit 143, and estimates the propagation path of the received signal using an interpolation method using linear interpolation, an interpolation method using a low-pass filter, or the like (step S1704). ). The division unit 145 receives the propagation path estimation value of the reception signal from the interpolation unit 144, performs complex division on the reception signal input from the FFT unit 13 by the propagation path estimation value, and outputs the equalized signal to the deframe processing unit 15 (Step S1705).

  Thus, the conventional propagation path estimation unit 14 estimates the propagation path of the received signal by calculating the SP carrier propagation path value and interpolating using the SP carrier propagation path value. The received signal is equalized using the propagation path estimated value of, and the equalized signal is output. Note that the propagation path value of the received signal may be estimated by calculating the propagation path value using a pilot signal other than the SP carrier and interpolating using the propagation path value of the pilot signal.

(Propagation path estimation process of Example 2)
Next, the process of the propagation path estimation unit 60 when the iterative decoding process from the second time to the predetermined number of times is performed will be described in detail. FIG. 10 is a block diagram illustrating the configuration of the propagation path estimation unit 60, FIG. 11 is a flowchart illustrating the process of the propagation path estimation unit 60, and FIG. 12 illustrates the process of the propagation path estimation unit 60. FIG.

  Referring to FIG. 10, this propagation path estimation unit 60 includes an SP generation unit 141, an SP extraction unit 142, division units 143, 164, and 167, a convolution decoding correct / incorrect information checking unit (inner code decoding correct / incorrect information checking unit) 161, A mapping point candidate selection unit 162, a final point selection unit (mapping point selection unit) 163, a synthesis unit 165, and an interpolation unit 166 are provided. Since the SP generation unit 141, the SP extraction unit 142, and the division unit 143 are the same as the processes performed by the conventional propagation path estimation unit 14 illustrated in FIGS.

  Referring to FIGS. 10 to 12, convolution decoding correct / incorrect information checking section 161 determines whether or not the convolution decoding correct / incorrect information input from OFDM framing section 39 is 5 (step S <b> 1101). If it is determined that the convolutional decoding correct / incorrect information is 5 (step S1101: Y), the test result indicating that the convolutional decoding correct / incorrect information is 5 is output to the mapping point candidate selecting unit 162, and the process proceeds to step 1102. On the other hand, if it is determined that the convolutional decoding correct / incorrect information is not 5 (step S1101: N), the process is terminated, and the convolutional decoding correct / incorrect information for the next decoding result is processed from step S1101.

  When the mapping point candidate selection unit 162 moves from step S1101 and inputs a check result indicating that the convolution decoding correct / incorrect information is 5 from the convolution decoding correct / incorrect information check unit 161, the convolution decoding correct / incorrect information is sent from the OFDM framing unit 39. A mapping point candidate (candidate point) of a predetermined length bit including the convolution decoding result input corresponding to is selected (step S1102). Then, the mapping point candidate selection unit 162 outputs the selected candidate points to the final point selection unit 163.

  As shown in FIG. 12, in the case of QPSK, four 2-bit long mapping points are arranged on the I and Q axes. For example, when the convolution decoding correct / incorrect information is 5 and the convolution decoding result is 1, the mapping point candidate selection unit 162 selects two of “11” and “01” from among the four mapping points. Select candidate points. Similarly, in b when the convolutional decoding correctness information is 5, when the convolutional decoding result is 0, two candidate points “00” and “01” are selected from the four mapping points. In the case of 64QAM, 64 6-bit mapping points are arranged on the I and Q axes. In this case, the mapping point candidate selection unit 162 uses the same process to obtain 64 mappings when there are two or more convolution decoding results in which the convolution decoding correct / incorrect information is 5 among the 6-bit convolution decoding results. Candidate points having bits of the same convolution decoding result value are selected from the points.

  The final point selection unit 163 receives candidate points from the mapping point candidate selection unit 162, receives a reception signal from the FFT unit 13, and receives the candidate point and a predetermined reception signal (a symbol and a symbol corresponding to the convolution decoding result in step S1102). The received signal at the carrier position) is compared with each mapping point, and the point with the shortest distance between the candidate point and the mapping point of the predetermined received signal (candidate point closest to the received signal) is final. It selects as a point (step S1103). Then, the final point selection unit 163 outputs the final point to the division unit 164.

  As illustrated in FIG. 12, the final point selection unit 163 performs processing of a in which two candidate points “11” and “01” are selected by the mapping point candidate selection unit 162, The received signal mapping points (α in FIG. 12) corresponding to the convolution decoding result of a are respectively compared, and the closest candidate point of “01” is selected as the final point. Similarly, in the process of b in which two candidate points “00” and “01” are selected by the mapping point candidate selection unit 162, these two candidate points and the received signal corresponding to the result of convolutional decoding of b are displayed. Each of the mapping points (β in FIG. 12) is compared, and the candidate point of “00” with the closest distance is selected as the final point.

  The division unit 164 receives the final point from the final point selection unit 163, complex-divides the predetermined reception signal (the reception signal in steps S1102 and S1103) input from the FFT unit 13 by the known signal of the final point, and performs post-division data γ is output to the synthesis unit 165 (step S1104). In this way, the value of the propagation path in the predetermined received signal is calculated as post-division data.

  The synthesizer 165 receives the post-division data γ (propagation path value in a predetermined received signal) from the division unit 164 and the post-division SP (SP propagation path value) from the division unit 143, respectively. Then, the post-division SP is combined and combined (step S1105). Then, the synthesis unit 165 outputs the predetermined reception signal and the propagation path value in the SP to the interpolation unit 166. Data carriers of predetermined received signals are arranged in one symbol according to the position of the convolution decoding result, and SP carriers are arranged in 12 symbols at intervals of 12 carriers. For this reason, the propagation path estimated value of the received signal is estimated based on the predetermined received signal and the value of the propagation path in the SP.

  The interpolation unit 166 receives the post-division data γ and the post-division SP (predetermined reception signal and propagation path value in the SP) from the synthesis unit 165, and estimates the propagation path of the reception signal, for example, by linear interpolation (step S1106). ). The division unit 167 receives the channel estimation value of the reception signal from the interpolation unit 166, performs complex division on the reception signal input from the FFT unit 13 by the channel estimation value, obtains an equalized signal, and deframes the unit 15 (Step S1107).

  Thus, when the first iterative decoding process (first decoding process) is performed, the propagation path estimation unit 60 performs the process by the conventional propagation path estimation unit 14 illustrated in FIGS. When iterative decoding processing is performed from the second time to a predetermined number of times, the mapping point of the received signal from the mapping point candidate (candidate point) including the convolutional decoding result (0, 1) when the convolutional decoding correctness information is 5 The closest candidate (final point) is selected, the value of the propagation path in the data carrier of the received signal is calculated, and interpolation is performed using the value of the propagation path in the data carrier of the received signal and the value of the propagation path in the SP carrier Thus, the propagation path of the reception signal is estimated, the reception signal is equalized using the propagation path estimation value of the reception signal, and the equalized signal is output.

  As described above, according to the receiving device 2 of the second embodiment illustrated in FIG. 9, the iterative decoding unit 42 performs the decoding result by RS decoding when iteratively decoding the concatenated code data of the RS code and the convolutional code. And the propagation path is estimated again based on decoding correct / incorrect information indicating whether or not the decoding result of RS decoding is correct, and a new equalized signal is calculated. Specifically, the RS decoding unit 22 generates a decoding result and decoding correctness information, and the OFDM framing unit 39 performs energy spreading, byte interleaving, convolutional coding, remapping, time interleaving, and frequency interleaved convolutional decoding. The result is subjected to OFDM framing, and then subjected to byte interleaving, convolutional coding, remapping, time interleaving, frequency interleaved convolutional decoding correct / failure information, OFDM framing, convolutional decoding result as a remapping signal and convolution The decryption correct / incorrect information is generated. Then, the propagation path estimation unit 60 uses the received signal and the convolutional decoding result when the convolutional decoding correct / incorrect information is 5, and uses the reception point and the convolutional decoding result as the candidate closest to the mapping point of the received signal from the mapping point candidate including the convolutional decoding result. To calculate a propagation path value in the received signal, obtain a new propagation path estimated value from the propagation path value and the SP propagation path value in the received signal, and obtain a new equalized signal. I made it. The iterative decoding unit 42 performs a decoding process using the new propagation path estimation value and the new equalized signal obtained in this way, generates video and audio data by a predetermined number of decoding processes, and outputs them to the outside.

  Thereby, the receiving device 2 can generate video / audio data with few errors. The iterative decoding unit 42 includes a conventional deframer 15, a frequency deinterleaver 16, a time deinterleaver 17, a demapping unit 18, a Viterbi decoder 19, a byte deinterleaver 20, an energy despreader 21, The RS decoding unit 22 is provided as it is, and the propagation path estimation unit 60 of the iterative decoding unit 42 includes the SP generation unit 141, the SP extraction unit 142, and the division unit 143 in the conventional propagation path estimation unit 14 as they are, and the propagation path estimation unit 60. Is newly provided with a convolution decoding correct / incorrect information checking unit 161 and the like. Thereby, it is possible to perform iterative decoding without correcting the decoding process such as the conventional Viterbi algorithm. Therefore, when iterative decoding is performed on the concatenated code data of the RS code and the convolutional code, the quality of the received signal can be improved with a simple configuration without modifying the conventional decoding process. That is, the reception area can be expanded and low C / N reception is possible.

  Next, a receiving apparatus according to the third embodiment will be described. In the third embodiment, when iterative decoding is performed on the concatenated code data of the RS code and the convolutional code, the soft decision value and the propagation path estimation value are changed by combining the first embodiment and the second embodiment.

  FIG. 13 is a block diagram illustrating a configuration of a receiving apparatus according to the third embodiment. The receiving device 3 includes a symbol synchronization unit 11, a GI removal unit 12, an FFT unit 13, and an iterative decoding unit 43. Since the symbol synchronization unit 11, the GI removal unit 12, and the FFT unit 13 are the same as those shown in FIGS. 1 and 9, the description thereof is omitted here.

[Iterative decoding unit]
The iterative decoding unit 43 is configured by combining the iterative decoding unit 41 illustrated in FIG. 1 and the iterative decoding unit 42 illustrated in FIG. 9. The iterative decoding unit 43, the deframing unit 15, and the frequency deinterleaver are combined. Unit 16, time deinterleaving unit 17, demapping unit 18, iterative decoding unit 41, byte interleaving units 31-2 and 34-2, convolutional code units 32-2 and 35-2, energy spreading unit 33-2, and remapping A unit 36, a time interleave unit 37, a frequency interleave unit 38, and an OFDM framing unit 39. The iterative decoding unit 41 has the same configuration as the iterative decoding unit 41 shown in FIG. 1, and includes a soft decision value changing unit 50, a Viterbi decoding unit 19, a byte deinterleaving unit 20, an energy despreading unit 21, and an RS decoding unit. 22, byte interleave units 31-1 and 34-1, convolutional code units 32-1 and 35-1, and an energy spreading unit 33-1 are provided. The components included in the iterative decoding unit 43 are the same as those shown in FIGS. 1 and 9, and a description thereof will be omitted here.

  The iterative decoding unit 43 receives the received signal from the FFT unit 13, calculates the equalized signal by estimating the propagation path of the received signal in the propagation path estimating unit 14, and calculates the soft decision value in the demapping unit 18. The iterative decoding unit 41 changes the soft decision value based on the decoding result of the video / audio data generated by the decoding process using the soft decision value and the decoding correctness information, and performs the decoding process using the changed soft decision value. The video / audio data is generated by repeating the change of the soft decision value and the decoding process a predetermined number of times. Further, the iterative decoding unit 43 feeds back the decoding result and decoding correctness information of the video / audio data generated by the iterative decoding unit 41, and the propagation path estimation unit 14 re-estimates the propagation path of the received signal to perform new equalization. The post-signal is calculated, the demapping unit 18 calculates a soft decision value using the new equalized signal, and the iterative decoding unit 41 performs the above-described processing to generate video / audio data. The iterative decoding unit 43 repeats such propagation path re-estimation, calculation of a new equalized signal, calculation of a soft decision value, and decoding processing in the iterative decoding unit 41 a predetermined number of times, thereby reducing video and audio data with few errors. Is generated.

  As described above, according to the receiving device 3 of the third embodiment illustrated in FIG. 13, the iterative decoding unit 43 performs the decoding result by RS decoding when performing iterative decoding on the concatenated code data of the RS code and the convolutional code. And repeating the decoding process by changing the soft decision value based on the decoding correct / incorrect information, and further estimating the propagation path again based on the decoding result and decoding correct / incorrect information obtained by repeating the decoding process by changing the soft decision value, A new post-equalization signal is calculated, a soft decision value is calculated, and the above decoding process is repeated. That is, the iterative decoding unit 43 performs decoding processing combining the first embodiment and the second embodiment, changes the soft decision value for the RS code and the concatenated code data of the convolutional code, repeats the decoding processing, and propagation. The process of estimating the path again and repeating the decoding process is performed. As a result, as in the first and second embodiments, the receiving device 3 can generate video / audio data with few errors. In addition, iterative decoding can be performed without correcting decoding processing such as the conventional Viterbi algorithm. Therefore, when iterative decoding is performed on the concatenated code data of the RS code and the convolutional code, the quality of the received signal can be improved with a simple configuration without modifying the conventional decoding process. That is, the reception area can be expanded and low C / N reception is possible.

  The present invention has been described with reference to the embodiments. However, the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the technical idea thereof. In the first to third embodiments, the Viterbi decoding unit 19 performs Viterbi decoding on the soft decision value. However, the present invention is not limited to Viterbi decoding. For example, other decoding methods such as Map decoding may be used. In short, any decoding method that decodes the soft decision value may be used. In the first to third embodiments, the receiving devices 1 to 3 perform decoding processing on the RS code and convolutional code concatenated code data. However, the present invention is limited to such concatenated code data. It is not a thing. For example, the present invention is also applied to concatenated code data of RS code and turbo code, and is also applied to concatenated code data of RS code and LDPC code (Low Density Parity-check Code).

1-44 101 Receiver 11 Symbol synchronization unit 12 GI removal unit 13 FFT unit 14 Channel estimation unit 15 Deframe unit 16 Frequency deinterleave unit 17 Time deinterleave unit 18 Demap unit 19 Viterbi decoding unit 20 Byte deinterleave Unit 21 energy despreading unit 22 RS decoding unit 31, 34, 73 byte interleaving unit 32, 35, 74 convolutional code unit 33, 72 energy spreading unit 36 remapping unit 37, 76 time interleaving unit 38, 77 frequency interleaving unit 39, 78 OFDM framing unit 41, 42, 43 Iterative decoding unit 50 Soft decision value changing unit 60 Channel estimation unit 71 RS coding unit 75 Mapping unit 79 IFFT unit 80 GI adding unit 100 Transmitting device 141 SP generating unit 142 SP extracting unit 143 , 145,164 167 divider 144,166 interpolation unit 161 convolutional decoding correctness information checking unit 162 mapping point candidate selection unit 163 final point selection unit 165 combining unit

Claims (6)

In a receiving apparatus that repeatedly performs a decoding process on a reception signal of concatenated code data including an outer code and an inner code,
An inner code decoding unit that performs a decoding process on the inner code using a soft decision value of the received signal;
The signal decoded by the inner code decoding unit is subjected to a decoding process on the outer code, and the decoding result corrected by the decoding process and the decoding correct / incorrect information indicating whether the decoding result is correct are generated. An encoding / decoding unit;
A correct / incorrect information encoding unit that performs an encoding process using the inner code on the decoding correct / incorrect information generated by the outer code decoding unit, and generates inner code decoded correct / incorrect information;
A decoding result encoding unit that performs an encoding process using the inner code on the decoding result generated by the outer code decoding unit, and generates an inner code decoding result;
Based on the inner code decoding correct / incorrect information generated by the correct / incorrect information encoding unit, the inner code decoding result generated by the decoding result encoding unit is set to a new soft decision value and used in the inner code decoding unit. A soft decision value changing unit for changing the soft decision value of the signal ,
The correct / incorrect information sign part is:
A delay unit that delays the decoding correct / incorrect information generated by the outer code decoding unit; and an adder that adds the decoding correct / incorrect information and the decoding correct / incorrect information delayed by the delay unit; Generated as inner code decoding correct / incorrect information,
The soft decision value changing unit
When the inner code decoding correct / incorrect information generated by the correct / incorrect information encoding unit is a predetermined value, when the inner code decoding result generated by the decoding result encoding unit is 1, a new soft decision value is set to 1, When the inner code decoding result is 0, a new soft decision value is set to 0, and the soft decision value of the received signal used in the inner code decoding unit is changed . In a receiving apparatus that repeatedly performs a decoding process on a reception signal of concatenated code data including an outer code and an inner code,
A pilot signal is extracted from the received signal, a propagation path of the extracted pilot signal is calculated using a pilot signal having a predetermined amplitude and phase, and propagation of the received signal is performed by interpolation processing using the pilot signal propagation path. A propagation path estimator that estimates a path, equalizes the received signal using a propagation path of the received signal, and generates an equalized signal;
A demapping unit that demaps the I and Q signals of the equalized signal generated by the propagation path estimation unit and generates a soft decision value;
An inner code decoding unit that performs a decoding process on the inner code using a soft decision value of the equalized signal generated by the demapping unit;
The signal decoded by the inner code decoding unit is subjected to a decoding process on the outer code, and the decoding result corrected by the decoding process and the decoding correct / incorrect information indicating whether the decoding result is correct are generated. An encoding / decoding unit;
A correct / incorrect information encoding unit that performs an encoding process using the inner code on the decoding correct / incorrect information generated by the outer code decoding unit, and generates inner code decoded correct / incorrect information;
A decoding result encoding unit that performs an encoding process using the inner code on the decoding result generated by the outer code decoding unit, and generates an inner code decoding result;
A mapping unit for mapping the inner code decoding correct / incorrect information generated by the correct / incorrect information encoding unit and the inner code decoding result generated by the decoding result encoding unit on the I and Q axes;
With
The propagation path estimator is
The inner code decoding correct / incorrect information indicating that the decoding result of the inner code decoding correct / incorrect information mapped by the mapping unit is correct is specified, and the specified inner code of the inner code decoding result mapped by the mapping unit is specified. Selects mapping point candidates on the I and Q axes including the inner code decoding result corresponding to the code decoding correct / incorrect information, and corresponds to the mapping point candidate and the inner code decoding correct / incorrect information indicating that the decoding result is correct Based on the predetermined received signal, a mapping point closest to the predetermined received signal is selected from the mapping point candidates, and the predetermined received signal is selected from the selected mapping point and the predetermined received signal. Calculate the propagation path, and perform interpolation processing using the pilot signal propagation path and the predetermined reception signal propagation path. A reception apparatus characterized by estimating a received signal propagation path, equalizing the received signal using the received signal propagation path, and generating a new equalized signal used in the demapping unit . The receiving device according to claim 2 ,
The correct / incorrect information sign part is:
A delay unit that delays the decoding correct / incorrect information generated by the outer code decoding unit, and an adder that adds the decoding correct / incorrect information and the decoding correct / incorrect information delayed by the delay unit, the addition result by the adder Is generated as inner code decoding correctness information,
The propagation path estimator is
An inner code decoding correct / incorrect information checking unit that checks whether the inner code decoding correct / incorrect information mapped by the mapping unit is information indicating that the decoding result is correct;
The inner code decoding correct / incorrect information, which is checked by the inner code decoding correct / incorrect information checking unit, is arranged on the I and Q axes including the inner code decoding result corresponding to the inner code decoding correct / incorrect information. A mapping point candidate selection unit for selecting mapping point candidates;
Based on a predetermined received signal corresponding to the inner code decoding correct / incorrect information indicating that the mapping point candidate selected by the mapping point candidate selector and the decoding result is correct, of the mapping point candidates, A mapping point selector for selecting a mapping point closest to the predetermined received signal;
A first division unit that complex-divides the predetermined reception signal by a mapping point selected by the mapping point selection unit, and calculates a propagation path of the predetermined reception signal;
An extractor for extracting a pilot signal from the received signal;
A second division unit for complex-dividing the pilot signal extracted by the extraction unit by a pilot signal having a predetermined amplitude and phase, and calculating a propagation path of the pilot signal;
A new propagation path of the reception signal is obtained by linear interpolation processing using the propagation path of the predetermined reception signal calculated by the first division unit and the propagation path of the pilot signal calculated by the second division unit. An interpolation unit to estimate;
A third division unit for complex-dividing the reception signal by a propagation path of the reception signal estimated by the interpolation unit to generate a new equalized signal;
A receiving apparatus comprising: The receiving device according to claim 2 ,
Further, a second correct / incorrect information encoding unit that performs an encoding process using the inner code on the decoding correct / incorrect information generated by the outer code decoding unit, and generates inner code decoded correct / incorrect information;
A decoding result generated by the outer code decoding unit, a second decoding result encoding unit that performs an encoding process using the inner code and generates an inner code decoding result; and
Based on the inner code decoding correct / incorrect information generated by the second correct / incorrect information encoding unit, the inner code decoding result generated by the second decoding result encoding unit is set to a new soft decision value, and the inner code A soft decision value changing unit for changing the soft decision value used in the decoding unit;
A receiving apparatus comprising: In a reception method in which decoding processing is repeatedly performed on a reception signal of concatenated code data including an outer code and an inner code,
A first step of calculating a soft decision value of the received signal;
A second step of performing a decoding process on the inner code using the soft decision value;
A third step of performing a decoding process on the outer code on the signal decoded in the second step, generating a decoding result obtained by performing error correction, and decoding correct / incorrect information indicating whether the decoding result is correct;
A fourth step of performing inner code decoding correct / incorrect information on the decoding correct / incorrect information generated in the third step,
A fifth step of performing an encoding process using the inner code on the decoding result generated in the third step and generating an inner code decoding result;
Based on the inner code decoding correct / incorrect information generated in the fourth step, the inner code decoding result generated in the fifth step is set as a new soft decision value and used in the second step. the has a sixth step of changing, the,
The fourth step includes
Delay the decoding correct / incorrect information generated in the third step, add the decoding correct / incorrect information and the delayed decoded correct / incorrect information, and generate the addition result as inner code decoding correct / incorrect information;
The sixth step includes
When the inner code decoding correct / incorrect information generated in the fourth step is a predetermined value, when the inner code decoding result generated in the fifth step is 1, a new soft decision value is set to 1, A reception method characterized by setting a new soft decision value to 0 when the result of code decoding is 0, and changing the soft decision value of the received signal in the first step . In a reception method in which decoding processing is repeatedly performed on a reception signal of concatenated code data including an outer code and an inner code,
A pilot signal is extracted from the received signal, a propagation path of the extracted pilot signal is calculated using a pilot signal having a predetermined amplitude and phase, and propagation of the received signal is performed by an interpolation process using the propagation path of the pilot signal. A first step of estimating a path, equalizing the received signal using a propagation path of the received signal, and generating an equalized signal;
A second step of demapping the equalized I and Q signals generated in the first step to generate a soft decision value;
A third step of performing a decoding process on the inner code using the soft decision value generated in the second step;
A decoding process for the outer code is performed on the signal decoded in the third step, and a decoding result corrected by the decoding process and decoding correct / incorrect information indicating whether the decoding result is correct are generated. And the steps
A fifth step of performing inner-code decoding correct / incorrect information by performing a code process using the inner code on the decoding correct / incorrect information generated in the fourth step;
A decoding step generated by the fourth step performs a coding process using the inner code and generates an inner code decoding result; and
A seventh step of mapping on the I and Q axes the inner code decoding correct / incorrect information generated by the fifth step and the inner code decoding result generated by the sixth step;
An eighth step of specifying inner code decoding correct / incorrect information indicating that the decoding result is correct among the inner code decoding correct / incorrect information mapped in the seventh step;
Mapping point candidates on the I and Q axes that include the inner code decoding result corresponding to the inner code decoding correct / incorrect information identified in the eighth step among the inner code decoding results mapped in the seventh step. A ninth step of selecting,
The mapping point candidate selected in the ninth step based on the mapping point candidate selected in the ninth step and the predetermined received signal corresponding to the inner code decoding correctness information specified in the eighth step A tenth step of selecting a mapping point closest to the predetermined received signal,
An eleventh step of calculating a propagation path of the predetermined received signal from the mapping point selected in the tenth step and the predetermined received signal;
A new propagation path of the reception signal is estimated by interpolation processing using the propagation path of the predetermined reception signal calculated in the eleventh step and the propagation path of the pilot signal, and is used in the second step. A twelfth step of generating an equalized signal;
A receiving method comprising:

Images