JP2010268509A - Interleave apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To obtain a communication apparatus that is applied to all communications using a multi-carrier modulation/demodulation system and a single-carrier modulation/demodulation system and further can greatly improve BER characteristics compared with the prior arts. <P>SOLUTION: A turbo coder includes a permutation means in which random sequences generated using prime numbers are disposed in buffers of N (natural number) rows × M (natural number) columns and the random sequences are used to permute bits on a row unit basis, thereby generating N kinds of random sequences. Further, a data sequence of an interleaver length is mapped to the N kinds of random sequences after permutation and then rows are switched in the data sequence after mapping according to prescribed rules, thereby generating a final permutation pattern. Finally, the generated permutation pattern is read out on a column unit basis. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a communication apparatus that employs a multicarrier modulation / demodulation method, and in particular, data communication using an existing communication line by a DMT (Discrete Multi Tone) modulation / demodulation method, an OFDM (Orthogonal Frequency Division Multiplex) modulation / demodulation method, or the like. The present invention relates to a communication device and a communication method that can realize the above. However, the present invention is not limited to a communication device that performs data communication by the DMT modulation / demodulation method, but can be applied to all communication devices that perform wired communication and wireless communication by a multicarrier modulation / demodulation method and a single carrier modulation / demodulation method via a normal communication line. Applicable.

  Hereinafter, a conventional communication apparatus will be described. For example, in wideband CDMA (W-CDMA: Code Division Multiple Access) using an SS (Spread Spectrum) scheme, a turbo code is proposed as an error correction code that greatly exceeds the performance of a convolutional code. This turbo code encodes an information bit sequence interleaved in parallel with a known encoded sequence, and is said to provide characteristics close to the Shannon limit. This is one of the correction codes. In the W-CDMA, the performance of the error correction code greatly affects the transmission characteristics in voice transmission and data transmission. Therefore, the transmission characteristics can be greatly improved by applying the turbo code.

  Here, operations of the transmission system and the reception system of the conventional communication apparatus using the turbo code will be specifically described. FIG. 19 is a diagram illustrating a configuration of a turbo encoder used in the transmission system. In FIG. 19A, 101 is a first recursive systematic convolutional encoder that convolutionally encodes an information bit sequence and outputs redundant bits, 102 is an interleaver, and 103 is replaced by the interleaver 102. This is a second recursive systematic convolutional encoder that convolutionally encodes the information bit sequence of the information and outputs redundant bits. FIG. 19B is a diagram illustrating an internal configuration of the first recursive systematic convolutional encoder 101 and the second recursive systematic convolutional encoder 103, and shows two recursive systematic convolutional encodings. Each encoder is an encoder that outputs only redundant bits. Further, the interleaver 102 used in the turbo encoder performs a process of randomly exchanging information bit sequences.

In the turbo encoder configured as described above, at the same time, the information bit sequence: x ₁ and the redundant bit sequence obtained by encoding the information bit sequence by the processing of the first recursive systematic convolutional encoder 101: x ₂ and redundant bit sequence: x _{3 obtained} by encoding the information bit sequence after the interleaving process by the process of the second recursive systematic convolutional encoder 103 is output.

FIG. 20 is a diagram illustrating a configuration of a turbo decoder used in the reception system. In FIG. 20, 111 is a first decoder that calculates a log likelihood ratio from received signal: y ₁ and received signal: y ₂ , 112 and 116 are adders, and 113 and 114 are interleavers. , 115 is a second decoder for calculating a log-likelihood ratio from the received signal: y ₁ and the received signal: y ₃ , 117 is a deinterleaver, and 118 is for determining the output of the second decoder 115. And a determinator that outputs an estimated value of the original information bit sequence. The received signals: y ₁ , y ₂ , and y ₃ are signals in which the information bit series: x ₁ and redundant bit series: x ₂ , x ₃ are affected by transmission line noise and fading, respectively.

In the turbo decoder configured as described above, first, the first decoder 111 performs log likelihood ratio of the estimated information bit: x _1k ′ estimated from the received signal: y _1k and the received signal: y _2k : L (x _1k ') for calculating a (k denotes the time). Here, the probability that the information bit: x _1k is 1 with respect to the probability that the information bit: x _1k is 0 is obtained. In the figure, Le (x _1k ) represents external information, and La (x _1k ) represents prior information that is the previous external information.

Next, the adder 112 calculates external information for the second decoder 115 from the log likelihood ratio as the calculation result. In the first decoding, since prior information is not obtained, La (x _1k ) = 0.

Next, the interleaver 113 and 114, the received signal: Y1k and the external information: Le a (x _1k), the received signal: in order to match the y ₃ times, rearranged signal. Thereafter, in the second decoder 115, similarly to the first decoder 111, the received signal: y ₁ , the received signal: y ₃ , and the previously calculated external information: Le (x _1k ). The log likelihood ratio: L (x _1k ′) is calculated. The adder 116 calculates external information: Le (x _1k ). At this time, the external information rearranged by the deinterleaver 117 is fed back to the first decoder 111 as prior information: La (x _1k ).

Finally, in the turbo decoder, the above process is repeatedly executed for a predetermined number of times to calculate a log likelihood ratio with higher accuracy, and the determiner 118 makes a determination based on the log likelihood ratio. To estimate the original information bit sequence. Specifically, for example, if the log likelihood ratio is “L (x _1k ′)> 0”, the estimated information bit: x _1k ′ is determined to be 1, and “L (x _1k ′) ≦ 0”. If so, the estimated information bit: x _1k ′ is determined to be 0.

  FIG. 21, FIG. 22, and FIG. 23 are diagrams showing the processing of the interleaver 102 used in the turbo encoder. Here, a process of randomly exchanging information bit sequences by the interleaver 102 will be described.

For example, in W-CDMA, a complex interleaver (hereinafter referred to as PIL) is generally used as an interleaver. This PIL has the following three characteristics.
(1) The rows and columns in the N (vertical axis: natural number) × M (horizontal axis: natural number) buffer are switched.
(2) A pseudo-random pattern using prime numbers is used in exchanging bits in a row.
(3) Avoid critical patterns by replacing rows.

Here, the operation of the conventional interleaver PIL will be described. For example, when the interleaver length: L _turbo = 512 bits, N = 10, M = P = 53 (L _turbo / N ≦ P + 1), primitive root: g ₀ = 2, the mapping pattern: c (i) It is created as shown in equation (1).

c (i) = (g0 * c (i-1)) mod P (1)

However, i = 1, 2,... (P-2) and c (0) = 1.

  Therefore, the mapping pattern C (i) is {1, 2, 4, 8, 16, 32, 11, 22, 44, 35, 17, 34, 15, 30, 7, 14, 28, 3, 6, 12 , 24, 48, 43, 33, 13, 26, 52, 51, 49, 45, 37, 21, 42, 31, 9, 18, 36, 19, 38, 23, 46, 39, 25, 50, 47 , 41, 29, 5, 10, 20, 40, 27}.

In PIL, the mapping pattern C (i) is skipped for each skip reading pattern: p _{PIP (j)} to perform bit replacement, and generate j rows of mapping patterns: C _j (i). To do. First, here, in order to obtain {p _{PIP (j)} }, {q _j (j = 1 to N−1)} is determined under the conditions of the following expressions (2), (3), and (4). .

q ₀ = 1 (2)
g. c. d {q _j , P-1} = 1 (where gcd is the greatest common divisor)
... (3)
q _j > 6, q _j > q _j−1 (where j = 1 to N−1) (4)

Therefore, {q _j } becomes {1, 7, 11, 13, 17, 19, 23, 29, 31, 37}, and {p _{PIP (j)} } becomes {37, 31, 29, 23, 19 , 17, 13, 11, 7, 1} (where PIP = N−1 to 0).

FIG. 21 is a diagram showing the result of skipping the mapping pattern C (i) based on the skip reading pattern: p _{PIP (j)} , that is, the result of rearranging each row in units of skip reading patterns.

Then, Figure 22, the mapping pattern after the rearrangement, the interleaver length: is a diagram showing a data arrangement in the case where the mapping data of the L _turbo = 512 bits. Here, data {0-52} is stored in the first line, data {53-105} is stored in the second line, data {106-158} is stored in the third line, and data {159-211} is stored in the fourth line. Data {212-264} on the 5th line, data {265-317} on the 6th line, data {318-370} on the 7th line, data {371-423} on the 8th line, 9 Data {424 to 476} is mapped to the line, and data {477 to 529} is mapped to the 10th line.

  Finally, FIG. 23 is a diagram showing a final rearrangement pattern. Here, according to a predetermined rule, the rows in the data array in FIG. 23 are replaced to generate a final rearrangement pattern (here, the order of the rows in the data array in FIG. 23 is reversed). In the PIL, the generated rearrangement pattern is read in units of columns, that is, vertically.

In this way, by using PIL as interleaving, it is possible to provide a turbo code that generates a codeword that has a good weight distribution in a wide range of interleaving lengths (for example, L _turbo = 257 to 8192 bits). .

FIG. 24 is a diagram showing BER (bit error rate) characteristics when a conventional turbo encoder and turbo decoder including the PIL are used. As shown in the figure, the BER characteristics improve as the SNR increases. For example, when the performance of a turbo code is determined using BER as shown in FIG. 24, “minimum hamming weight: w _min ” after turbo code affects the BER of high SNR. Specifically, it is generally known that when the minimum hamming weight is small, the BER of the error floor area (area where the BER falls slowly) increases.

The minimum hamming weight means, for example, the minimum value of the number of “1” in each pattern that can be taken by the series (x ₁ , x ₂ , x ₃ ) shown in FIG. So, for example,
x ₁ =… 00100,000…
x ₂ = ... 00010100000 ...
x ₃ = ... 00010101000 ...
Is a pattern indicating a minimum value of the number of “1”, the minimum hamming weight of this turbo encoder is w _min = 7. Here, x ₁ and x ₂ represent the input data series of the encoder, and x ₃ represents the output data series from the encoder.

  As described above, in the conventional communication device, by applying a turbo code as an error correction code, even when the distance between signal points becomes short according to the multi-level modulation method, voice transmission or data Transmission characteristics in transmission can be greatly improved, and characteristics superior to known convolutional codes have been obtained.

  Further, in the conventional communication apparatus, turbo coding is performed on all input information sequences (if there are a plurality of information bit sequences, all of the sequences), and further on the receiving side. All the encoded signals are turbo-decoded and then soft-decision is performed. More specifically, for example, all 16-bit data (0000 to 1111: 4-bit constellation) is determined for 16QAM, and all 8-bit data is determined for 256QAM. become.

  However, in the conventional communication apparatus employing the turbo code, for example, an encoder (corresponding to a recursive systematic convolutional encoder) and an interleaver used in the conventional turbo encoder shown in FIG. However, there is a problem that turbo coding using such a conventional encoder and interleaver cannot obtain optimum transmission characteristics close to the Shannon limit, that is, optimum BER characteristics. there were.

  The conventional turbo encoder is specialized for one information bit sequence and has a problem that it does not support two information bit sequences.

  The present invention has been made in view of the above, and can be applied to all communications using a multicarrier modulation / demodulation method and a single carrier modulation / demodulation method. Further, the BER characteristics can be greatly improved as compared with the prior art. It is an object of the present invention to obtain a feasible communication device and communication method.

  In order to solve the above-described problems and achieve the object, in the communication apparatus according to the present invention, the turbo encoder uses a prime number in a buffer of N (natural number) rows × M (natural number) columns. By arranging the generated random sequences and rearranging bits in units of rows using the random sequences, N types of random sequences are generated, and an interleaver length is added to the N types of random sequences after the rearrangement. The data sequence is mapped, and the rearrangement between the rows in the data series after the mapping is performed according to a predetermined rule to generate a final rearrangement pattern, and the generated rearrangement pattern is converted to a column unit. Rearrangement means for reading out the data.

  In the communication apparatus according to the next invention, the turbo encoder arranges a random sequence generated using prime numbers in a buffer of N (natural number) rows × M (natural number) columns, and in units of rows. By shifting the random sequence one column at a time, N types of random sequences are generated, and an interleaver length data sequence is mapped to the N types of random sequences after the shift. By performing the replacement according to a predetermined rule, a final rearrangement pattern is generated, and rearrangement means for reading out the generated rearrangement pattern in units of columns is provided.

  In the communication apparatus according to the next invention, when two information bit sequences are input to the turbo encoder, the rearranging means prevents the distance between signal points of the two information bit sequences from becoming zero. , Each is rearranged.

  In the communication method according to the next invention, a random sequence generated using a prime number is arranged in a buffer of N (natural number) rows × M (natural number) columns, and the random sequence is used to set a bit in a row unit. A random sequence generating step for generating N types of random sequences, a mapping step for mapping a data sequence of an interleaver length to the N types of random sequences after the rearrangement, A rearrangement pattern generation step for generating a final rearrangement pattern by performing replacement between rows in the data series according to a predetermined rule, and a reading step for reading out the generated rearrangement pattern in units of columns. , Including.

  In the communication method according to the next invention, a random sequence generated using prime numbers is arranged in a buffer of N (natural number) rows × M (natural number) columns, and the random sequence is arranged in one column for each row. In the random sequence generation step for generating N types of random sequences, a mapping step for mapping a data sequence of an interleaver length to the N types of random sequences after the shift, A rearrangement pattern generation step for generating a final rearrangement pattern by performing inter-row replacement according to a predetermined rule, and a reading step for reading out the generated rearrangement pattern in units of columns. It is characterized by that.

  In the communication method according to the next invention, in addition, when two information bit sequences are input to the turbo encoder, the distance between signal points of the two information bit sequences is not zero. It is characterized by sorting each.

  According to the present invention, it is possible to improve randomness while maintaining the same distance between signal points as in the past by using a rearrangement unit that rearranges bits in units of rows using a random sequence. The error correction capability can be improved. As a result, the demodulation characteristic on the receiving side can be greatly improved, and an effect is obtained that a communication apparatus that can realize an optimum transmission characteristic close to the Shannon limit, that is, an optimum BER characteristic can be obtained.

  According to the next invention, by using the rearranging means for shifting the random sequence one column at a time in units of rows, the distance between signal points can be greatly improved as compared with the prior art. It becomes possible to further improve the error correction capability by the combination. As a result, the demodulation characteristics on the receiving side can be further greatly improved, so that an optimum transmission characteristic close to the Shannon limit, that is, a communication device capable of realizing the optimum BER characteristic can be obtained. .

  According to the next invention, for example, by obtaining all the distances between the information bit series: U1 and the information bit series: U2, and further obtaining all the inter-signal distances for each distance, the optimum transmission line characteristics can be obtained. Select the resulting sorting means. Accordingly, a communication device capable of realizing a rearrangement unit optimal for a turbo encoder having two information bit sequences, that is, a rearrangement unit that can sufficiently take a distance between signal points between information bit sequences. There is an effect that it can be obtained.

  According to the next invention, by using a random sequence generation step of rearranging bits in units of rows using a random sequence, it is possible to improve randomness while maintaining a signal point distance equivalent to the conventional one. Therefore, the error correction capability can be improved. As a result, the demodulation characteristic on the receiving side can be greatly improved, and an effect is obtained that a communication method that can realize an optimum transmission characteristic close to the Shannon limit, that is, an optimum BER characteristic can be obtained.

  According to the next invention, by using the random sequence generation step that shifts the random sequence one column at a time in units of rows, the distance between signal points can be greatly improved as compared with the prior art. The error correction capability can be further improved by combining with the above. As a result, the demodulation characteristic on the receiving side can be further greatly improved, so that an optimum transmission characteristic close to the Shannon limit, that is, a communication method capable of realizing the optimum BER characteristic can be obtained. .

  According to the next invention, for example, by obtaining all the distances between the information bit series: U1 and the information bit series: U2, and further obtaining all the inter-signal distances for each distance, the optimum transmission line characteristics can be obtained. Select the resulting interleaver. As a result, a communication method capable of realizing rearrangement optimal for a turbo encoder having two information bit sequences, that is, rearrangement that can sufficiently take the distance between signal points between information bit sequences is obtained. There is an effect that can be.

FIG. 1 is a diagram showing a configuration of an encoder and a decoder used in a communication apparatus according to the present invention. FIG. 2 is a diagram showing the configuration of the transmission system of the communication apparatus according to the present invention. FIG. 3 is a diagram showing the configuration of the reception system of the communication apparatus according to the present invention. FIG. 4 is a diagram showing signal point arrangements for various digital modulations. FIG. 5 is a diagram illustrating a configuration of the turbo encoder 1. FIG. 6 is a diagram illustrating an example of a recursive systematic convolutional encoder that configures the same code as the recursive systematic convolutional encoder of FIG. FIG. 7 is a diagram showing BER characteristics when transmission data is decoded using the turbo encoder of the present invention, and BER characteristics when transmission data is decoded using a conventional turbo encoder. FIG. 8 is a diagram illustrating the minimum hamming weight of the turbo encoder of the present invention and the minimum hamming weight of the conventional turbo encoder when a specific interleaver size is employed. FIG. 9 is a diagram showing processing of the first embodiment of the interleaver used in the turbo encoder according to the present invention. FIG. 10 is a diagram showing processing of the first embodiment of the interleaver used in the turbo encoder according to the present invention. FIG. 11 is a diagram showing processing of the first embodiment of the interleaver used in the turbo encoder according to the present invention. FIG. 12 is a diagram in which the PPI of the first embodiment and the conventional PIL are quantitatively compared. FIG. 13 is a diagram showing the processing of the second embodiment of the interleaver used in the turbo encoder according to the present invention. FIG. 14 is a diagram showing the processing of the second embodiment of the interleaver used in the turbo encoder according to the present invention. FIG. 15 is a diagram illustrating processing of the second embodiment of the interleaver used in the turbo encoder according to the present invention. FIG. 16 is a diagram in which the PPI of the second embodiment and the conventional PIL are quantitatively compared. FIG. 17 is a diagram illustrating the distance between signal points U1 and U2 when the distance between the information bit sequence U1 and the information bit sequence U2 is five rows. FIG. 18 is a diagram illustrating optimum values of two interleavers when rearrangement is performed under the same conditions as in the first embodiment. FIG. 19 is a diagram showing a configuration of a conventional turbo encoder used in a transmission system. FIG. 20 is a diagram showing a configuration of a conventional turbo decoder used in the reception system. FIG. 21 is a diagram showing interleaver processing used in a conventional turbo encoder. FIG. 22 is a diagram showing interleaver processing used in a conventional turbo encoder. FIG. 23 is a diagram illustrating interleaver processing used in a conventional turbo encoder. FIG. 24 is a diagram showing bit error rate characteristics when a conventional turbo encoder and turbo decoder are used.

  Embodiments of a communication apparatus according to the present invention will be described below in detail with reference to the drawings. Note that the present invention is not limited to the embodiments.

Embodiment 1 FIG.
FIG. 1 shows a configuration of an encoder (turbo encoder) and a decoder (a combination of a turbo decoder, a hard discriminator, and an R / S (Reed-Solomon code) decoder) used in a communication apparatus according to the present invention. Specifically, FIG. 1 (a) is a diagram showing a configuration of an encoder in the present embodiment, and FIG. 1 (b) is a diagram showing a configuration of a decoder in the present embodiment.

  In the communication apparatus according to the present embodiment, both the encoder and the decoder are provided, and by having a highly accurate data error correction capability, excellent transmission characteristics in data communication and voice communication are obtained. And In the present embodiment, for convenience of explanation, both the above configurations are provided. However, for example, a transmitter including only two encoders may be assumed, while a decoder is provided. It is good also as a receiver provided only with this.

  In the encoder shown in FIG. 1A, reference numeral 1 denotes a turbo encoder that can obtain performance close to the Shannon limit by adopting a turbo code as an error correction code. For example, in the turbo encoder 1, 2 bits of information bits and 2 bits of redundant bits are output in response to the input of 2 bits of information bits. Further, here, in order to make the correction capability for each information bit uniform on the receiving side, Generate bits.

On the other hand, in the decoder of FIG. 1B, reference numeral 11 denotes _a first decoder that calculates a log likelihood ratio from _a received signal: Lcy (corresponding to received signals: y ₂ , y ₁ , and ya described later). , 12 and 16 are adders, 13 and 14 are interleavers, and 15 is a logarithmic likelihood ratio calculated from a received signal: Lcy (corresponding to received signals: y ₂ , y ₁ , y _b described later). 2 is a decoder, 17 is a deinterleaver, 18 is a first determiner that determines the output of the first decoder 15 and outputs an estimated value of the original information bit sequence, and 19 is a read A first R / S decoder that decodes the Solomon code and outputs a more accurate information bit sequence. 20 determines the output of the second decoder 15 and outputs an estimated value of the original information bit sequence 21 is a second determinator that decodes the Reed-Solomon code. Is a second R / S decoder that outputs a highly accurate information bit sequence, and 22 is a hard decision of Lcy (corresponding to received signals to be described later: y ₃ , y ₄ ...) To estimate the original information bit sequence. It is the 3rd determination device which outputs a value.

  Here, before describing the operations of the encoder and decoder, the basic operation of the communication apparatus according to the present invention will be briefly described with reference to the drawings. For example, as a wired digital communication system that performs data communication using a DMT (Discrete Multi Tone) modulation / demodulation system, an ADSL (Asymmetric Digital Subscriber Line) that performs high-speed digital communication of several megabits / second using an existing telephone line is used. ) And xDSL communication methods such as a high-bit-rate digital subscriber line (HDSL) communication method. This method is standardized in ANSI T1.413. Hereinafter, for the description of the present embodiment, for example, a communication apparatus applicable to the ADSL is used.

  FIG. 2 is a diagram showing the configuration of the transmission system of the communication apparatus according to the present invention. In FIG. 2, in the transmission system, transmission data is multiplexed by a multiplex / sync control (corresponding to MUX / SYNC CONTROL shown in the figure) 41, and cyclic redundancy check (CRC: Cyclic check) is performed on the multiplexed transmission data. Error detection codes are added at 42 and 43 (corresponding to redundancy check), and FEC codes are added and scrambled at forward error correction (corresponding to SCRAM & FEC) 44 and 45.

  There are two paths from the multiplex / sync control 41 to the tone ordering 49, one is an interleaved data buffer path including an interleave 46, and the other. Is a fast data buffer path that does not include interleaving, and here, the delay of the interleaved data buffer path for performing the interleaving process becomes larger.

  Thereafter, the transmission data is subjected to rate conversion processing at rate converters (corresponding to RATE-CONVERTOR) 47 and 48, and tone ordering processing is performed at tone ordering (corresponding to TONE ORDERRING) 49. Based on the transmission data after the tone ordering process, constellation encoder / gain scaling (corresponding to CONSTELLATION AND GAIN SCALLNG) 50 creates constellation data, which is converted into an inverse fast Fourier transform (IFFT: Inverse Fast Fourier transform). Equivalent) 51, the inverse fast Fourier transform is performed.

  Finally, the input parallel / serial buffer (equivalent to INPUT PARALLEL / SERIAL BUFFER) 52 converts the parallel data after Fourier transform into serial data, and the analog processing / digital-analog converter (equivalent to ANALOG PROCESSING AND DAC) 53 The digital waveform is converted into an analog waveform, and after executing the filtering process, the transmission data is transmitted on the telephone line.

  FIG. 3 is a diagram showing the configuration of the reception system of the communication apparatus according to the present invention. In FIG. 3, the receiving system performs filtering processing on received data (the above-mentioned transmission data) by an analog processing / analog-digital converter (corresponding to ANALOG PROCESSING AND ADC shown in the figure) 141, and then converts the analog waveform into a digital waveform. Time domain equalizer (corresponding to TEQ) 142 performs time domain adaptive equalization processing.

  Data subjected to time domain adaptive equalization processing is converted from serial data to parallel data in an input serial / parallel buffer (equivalent to INPUT SERIAL / PARALLEL BUFFER) 143, and fast Fourier transform is performed on the parallel data. (FFT: equivalent to Fast Fourier transform) 144 performs fast Fourier transform, and then frequency domain equalizer (corresponding to FEQ) 145 performs frequency domain adaptive equalization.

  The data subjected to frequency domain adaptive equalization processing is decoded by a constellation decoder / gain scaling (equivalent to CONSTELLATION DECODER AND GAIN SCALLNG) 146 and tone ordering (equivalent to TONE ORDERRING) 147 ( It is converted into serial data by the maximum likelihood decoding method) and tone ordering processing. After that, rate conversion processing (corresponding to RATE-CONVERTOR) 148, 149, deinterleaving (corresponding to DEINTERLEAVE) 150, deinterleaving processing, forward error correction (corresponding to DESCRAM & FEC) 151, 152 FEC processing and descrambling processing , And cyclic redundancy check (equivalent to cyclic redundancy check) 153, 154 and other processing such as cyclic redundancy check are performed, and finally the received data is reproduced from multiplex / sync control (equivalent to MUX / SYNC CONTROL) 155 Is done.

  In the communication apparatus as described above, each of the reception system and the transmission system has two paths, and by using these two paths properly or by operating these two paths simultaneously, a low transmission delay and a high rate are achieved. Data communication can be realized.

  In the communication apparatus configured as described above, the encoder shown in FIG. 1A is positioned in the constellation encoder / gain scaling 50 in the transmission system, and the decoder shown in FIG. , The constellation decoder / gain scaling 146 in the receiving system.

  Hereinafter, operations of the encoder (transmission system) and the decoder (reception system) in the present embodiment will be described in detail with reference to the drawings. First, the operation of the encoder shown in FIG. In the present embodiment, for example, a 16QAM system is adopted as multi-level quadrature amplitude modulation (QAM). Also, in the encoder according to the present embodiment, unlike the conventional technique in which turbo coding is performed on all input data (4 bits), as shown in FIG. Turbo coding is performed only for the other bits, and the input data is output as it is for the other upper bits.

  Here, the reason why turbo coding is performed only on the lower two bits of input data will be described. 4A and 4B are diagrams showing signal point arrangements for various digital modulations. Specifically, FIG. 4A is a signal phase arrangement of a 4-phase PSK (Phase Shift Keying) system, and FIG. 4B is a 16QAM system. This is a signal point arrangement, and (c) is a signal point arrangement of the 64QAM system.

  For example, in all the modulation scheme signal point arrangements described above, when the received signal point is at the position a or b, the receiving side normally estimates the most probable data as an information bit sequence (transmission data) by soft decision. . That is, the signal point that is the closest to the reception signal point is determined as transmission data. However, at this time, for example, focusing on the reception signal points a and b in FIG. 4, in any case (corresponding to FIGS. 4 (a), (b), and (c)) It can be seen that the lower 2 bits are (0, 0) (0, 1) (1, 0) (1, 1). Therefore, in the present embodiment, a turbo having an excellent error correction capability with respect to the lower 2 bits of four signal points (that is, the four points having the closest distance between signal points) whose characteristics may be deteriorated. Encoding is performed, and soft decision is performed on the receiving side. On the other hand, the other high-order bits whose characteristics are unlikely to deteriorate are output as they are, and a hard decision is made on the receiving side.

  Thereby, in the present embodiment, it is possible to improve the characteristics that may be deteriorated along with the multi-level, and furthermore, because turbo encoding is performed only on the lower 2 bits of the transmission signal, Compared with the prior art in which all bits are the target of turbo coding, the amount of calculation can be greatly reduced.

Next, the operation of the turbo encoder 1 shown in FIG. 1A, in which turbo encoding is performed on the input transmission data of lower 2 bits: u ₁ and u ₂ will be described. For example, FIG. 5 is a diagram showing a configuration of the turbo encoder 1, and more specifically, FIG. 5 (a) is a diagram showing a block configuration of the turbo encoder 1, and FIG. 5 (b) is a recursive organization. It is a figure which shows an example of the circuit structure of a convolutional encoder. Here, the configuration shown in FIG. 5B is used as the recursive systematic convolutional encoder, but the present invention is not limited to this. For example, the same recursive systematic convolutional encoder as in the prior art or other known ones. The recursive systematic convolutional encoder may be used.

In FIG. 5A, reference numeral 31 denotes _a first recursive systematic convolutional encoder that convolutionally encodes transmission data: u ₁ and u ₂ corresponding to an information bit sequence and outputs redundant data: u _a . , 32 and 33 are interleavers, 34 interleave processed data: u _1t, u _2t convolutional encoding to redundant data: a second recursive systematic convolutional encoder that outputs a u _b. The turbo encoder 1 simultaneously transmits transmission data: u ₁ , u ₂ , redundant data: u _a obtained by encoding the transmission data by the processing of the first recursive systematic convolutional encoder 31, and the second recursive tissue by treatment of convolutional encoder 34 the data after interleaving and coding (time different from other data) redundant data outputs and u _b, a.

In the recursive systematic convolutional encoder shown in FIG. 5B, 61, 62, 63 and 64 are delay units, and 65, 66, 67, 68 and 69 are adders. In this recursive systematic convolutional encoder, an adder 65 at the first stage receives input transmission data: u ₂ (or data: u _1t ) and fed back redundant data: u _a (or redundant data: u _b) and the sum output, the second stage of the adder 66, the transmission data is input: u ₁ (or the data: u _2t) and outputs adds the output of the delay unit 61, the addition of the third stage The unit 67 adds and outputs the input transmission data: u ₁ (or data: u _2t ), transmission data: u ₂ (or data: u _1t ), and the output of the delay unit 62, and adds the fourth stage adder 68 is input transmission data: u ₁ (or data: u _2t ), transmission data: u ₂ (or data: u _1t ), the output of the delay unit 63, and redundant data fed back: u _a (or redundant data). : u _b) and the sum output of the final-stage adder 6 But the transmission data are input: u ₂ (or the data: u _1t) adds the output of a delay unit 64, and finally redundant data: u _a (redundant data: u _b) outputs a.

In the turbo encoder 1, the weights in the redundant bits are biased so that the estimation accuracy of the transmission data: u ₁ and u ₂ on the receiving side using the redundant data: u _a and u _b is uniform. It does not occur. That is, in order to make the estimation accuracy of the transmission data: u ₁ and u ₂ uniform, for example, the transmission data: u ₂ is added to the adders 65, 67, 68, in the first recursive systematic convolutional encoder 31. 69 (see FIG. 5B) and interleaved data: u _2t is input to adders 66 to 68 in the second recursive systematic convolutional encoder 34, while transmission data: u ₁ is input to adders 66 to 68 in the first recursive systematic convolutional encoder 31, and the interleaved data: u _1t is added in the second recursive systematic convolutional encoder 34. By inputting the data into the units 65, 67, 68, and 69, the number of delay units that pass until the output is made the same between the transmission data: u ₁ series and the transmission data: u ₂ series.

As described above, when the encoder shown in FIG. 1A is used, as an interleaving effect, it is possible to improve error correction capability against bursty data errors, and further transmit data: The input of the sequence of u ₁ and the transmission data: the input of the sequence of u ₂ are exchanged between the first recursive systematic convolutional encoder 31 and the second recursive systematic convolutional encoder 34. This makes it possible to make the estimation accuracy of the transmission data u ₁ and u ₂ on the receiving side uniform.

  FIG. 6 is a diagram illustrating an example of a recursive systematic convolutional encoder that configures the same code as the recursive systematic convolutional encoder of FIG. Therefore, even when the recursive systematic convolutional encoder shown in FIG. 5B is replaced with the circuit configuration shown in FIG. 6, the same effect as described above can be obtained.

In the recursive systematic convolutional encoder shown in FIG. 6, reference numerals 71, 72, 73, and 74 are delay units, and reference numerals 75, 76, 77, and 78 are adders. In this recursive systematic convolutional encoder, the first stage adder 75 adds and outputs the input transmission data: u ₁ (or data: u _2t ) and the output of the delay unit 71, and the second stage. The adder 76 adds and outputs the input transmission data: u ₁ (or data: u _2t ), the transmission data: u ₂ (or data: u _1t ), and the output of the delay unit 72, and outputs the third stage. The adder 77 adds and outputs the input transmission data: u ₁ (or data: u _2t ), the output of the delay unit 73 and the output of the delayed delay unit 74, and the adder 78 in the final stage inputs Transmission data: u ₂ (or data: u _1t ) and the output of the delay unit 74 are added, and finally redundant data: u _a (redundant data: u _b ) is output.

Next, the operation of the decoder shown in FIG. In the present embodiment, a case will be described in which, for example, a 16 QAM system is adopted as multi-level quadrature amplitude modulation (QAM). Also, in the decoder of the present embodiment, turbo decoding is performed on the lower 2 bits of the received data, the original transmission data is estimated by soft decision, and the received data is transferred to the third bit for the other higher bits. The original transmission data is estimated by making a hard decision with the determiner 22. However, the received signal _{Lcy: y 4, y 3,} y 2, y 1, y a, y b , the output of each of the transmitting _{side: u 4, u 3, u} 2, u 1, u a, the u _b This is a signal that is affected by noise and fading in the transmission path.

First, the received signal Lcy: In y _2, y _1, y _a, turbo decoder receives a y _b, the first decoder 11, the received signal Lcy: extracts y _2, y _1, y _a, these Information bits estimated from received signals (original transmission data: corresponding to u _1k , u _2k ): log likelihood ratios of u _1k ′, u _2k ′: L (u _1k ′), L (u _2k ′) (K represents time). In other words, in this case, and thus to determine the probability u _2k is 1 for probability u _2k is 0, the probability u _1k is 1 for probability u _1k is 0, the. In the following description, u _1k and u _2k are simply referred to as u _k, and u _1k ′ and u _2k ′ are simply referred to as u _k ′.

However, in FIG. 1 (b), Le (u k) represents external information, La (u _k) represents the prior information is previous external information. As the decoder for calculating the log likelihood ratio, for example, a known maximum a posteriori decoder (MAP algorithm: Maximum A-Posteriori) is often used. For example, a known Viterbi decoder is used. It is good.

Next, the adder 12 calculates external information: Le (u _k ) for the second decoder 15 from the log likelihood ratio as the calculation result. However, in the first decoding, for the prior information is not being sought, it is a La (u _k) = 0.

Next, the interleavers 13 and 14 perform signal rearrangement on the received signal Lcy and the external information: Le (u _k ). Then, in the second decoder 15, similarly to the first decoder 11, based on the received signal Lcy and the previously calculated prior information: La (u _k ), the log likelihood ratio: L (U _k ′) is calculated.

Thereafter, the adder 16 calculates the external information: Le (u _k ) as in the adder 12. At this time, the external information rearranged by the deinterleaver 17 is fed back to the first decoder 11 as prior information: La (u _k ).

In the turbo decoder, the above process is repeatedly performed a predetermined number of times (the number of iterations) to calculate a log likelihood ratio with higher accuracy, and the first determiner 18 and the second The determiner 20 determines a signal based on the log likelihood ratio and estimates the original transmission data. More specifically, for example, if the log likelihood ratio is “L (u _k ′)> 0”, the estimated information bit: u _k ′ is determined as 1, and “L (u _k ′) ≦ 0”. If so, the estimated information bit: u _k ′ is determined to be 0. The received signals Lcy: y ₃ , y ₄ ... Received simultaneously are hard-decided using the third determiner 22.

  Finally, the first R / S decoder 19 and the second R / S decoder 21 check errors using a Reed-Solomon code by a predetermined method, and determine that the estimation accuracy exceeds a certain standard. At the stage, the above repeating process is terminated. Then, using the Reed-Solomon code, the error correction of the estimated original transmission data is performed by each determiner, and transmission data with higher estimation accuracy is output.

  Here, the original transmission data estimation method by the first R / S decoder 19 and the second R / S decoder 21 will be described according to a specific example. Here, three methods are given as specific examples. As the first method, for example, every time the original transmission data is estimated by the first determiner 18 or the second determiner 20, the corresponding first R / S decoder 19 or second R / S decoder 21 alternately checks for errors, and when one of the R / S decoders determines that there is no error, the above repetitive processing by the turbo encoder is terminated, and Reed-Solomon Error correction is performed on the estimated original transmission data using a code, and transmission data with higher estimation accuracy is output.

  Further, as the second method, each time the original transmission data is estimated by the first determiner 18 or the second determiner 20, the corresponding first R / S decoder 19 or second R / S decoder 21 alternately checks for errors, and when both R / S decoders determine that there is no error, the repetition process by the turbo encoder is terminated, and the Reed-Solomon code is Using this, error correction is performed on the estimated original transmission data, and transmission data with higher estimation accuracy is output.

  In addition, the third method improves the problem of erroneous correction when it is erroneously determined that there is no error in the first and second methods and the repetitive processing is not performed. For example, after performing a predetermined number of iterations determined in advance and reducing the bit error rate to some extent, performing an error correction of the estimated transmission data using a Reed-Solomon code, Output transmission data with higher estimation accuracy.

  As described above, when the decoder shown in FIG. 1B is used, even when the constellation increases as the modulation scheme becomes multi-valued, the lower 2 of the received signal with the possibility of characteristic degradation. Soft decision processing with a large amount of calculation by including a turbo decoder that performs soft decision processing on bits and error correction using a Reed-Solomon code, and a decision unit that performs hard decision on other bits in the received signal Reduction and good transmission characteristics can be realized.

  Further, by estimating transmission data using the first R / S decoder 19 and the second R / S decoder 21, the number of iterations can be reduced, and a soft decision process with a large amount of calculation and its processing time Can be further reduced. Note that, in a transmission path in which random errors and burst errors are mixed, excellent transmission characteristics can be obtained by using in combination with an RS code (Reed Solomon) that performs error correction in symbol units or other known error correction codes. Is generally known to be obtained.

Here, the BER (bit error rate) characteristics when the transmission data is decoded using the turbo encoder of the present invention and the BER characteristics when the transmission data is decoded using the conventional turbo encoder are compared. . FIG. 7 is a diagram showing both BER characteristics. For example, when judging the performance of a turbo code using BER, the “minimum hamming weight: w _min ” after turbo coding affects the BER of high SNR. That is, it is generally known that when the minimum hamming weight is small, the BER of the error floor area (area where the BER falls gradually) increases. Thus, it can be seen that, in the high E _b / N _o region, that is, the error floor region, the minimum hamming weight: w _min most affects the BER characteristics. Therefore, here, the minimum hamming weight of the turbo codeword is adopted as an index for comparing the performance of each encoder.

  FIG. 8 is a diagram illustrating the minimum hamming weight of the turbo encoder of the present invention and the minimum hamming weight of the conventional turbo encoder when a specific interleaver is employed. The minimum hamming weight is obtained by turbo-coding the input information bit sequence having a hamming weight of '2' and '3' over all patterns, and then obtaining the humming weight of the encoded sequence. Is the minimum value.

  From the comparative examination results in FIG. 7 and FIG. 8, it can be said that the performance of the turbo encoder shown in FIG. 1 having a large minimum hamming weight and a low BER characteristic in the error floor region is clearly superior to the prior art.

  In this way, the recursive systematic convolutional encoder (encoder) used in the turbo encoder 1 is supplied with either one of the transmission data series as shown in FIG. 5B and FIG. By adopting a form that is input to the adder, the influence of the transmission data can be more strongly reflected on the redundant data. That is, the demodulation characteristic on the receiving side can be greatly improved as compared with the prior art.

  As described above, in the conventional turbo encoder and the turbo encoder shown in FIG. 1, it is assumed that the same interleaver is used, and due to the difference between recursive systematic convolutional encoders, Improved the demodulation characteristics. In the following description, by using the interleaver according to the present embodiment, the demodulation characteristic on the receiving side is further greatly improved, and the optimum transmission characteristic close to the Shannon limit, that is, the optimum BER characteristic is obtained.

  FIGS. 9, 10, and 11 are diagrams showing processing of the first embodiment of the interleavers 32 and 33 used in the turbo encoder shown in FIG. Here, a process of randomly exchanging information bit sequences using interleavers 32 and 33 according to the present embodiment will be described.

For example, in W-CDMA, PIL is generally used as an interleaver. Here, instead of this PIL, a parallel prime interleaver (hereinafter referred to as PPI) which is an interleaver according to the present invention. Will be used. This PPI has the following three features.
(1) The rows and columns in the N (vertical axis: natural number) × M (horizontal axis: natural number) buffer are switched.
(2) The mapping pattern is shifted by (g ₀ × c (i−1)) mod P in units of rows.
(3) Avoid critical patterns by replacing rows.

Here, the operation of the PPI that is the interleaver of the present embodiment will be described. In the present embodiment, rearrangement is performed under the same conditions as the interleaver described in the prior art in order to evaluate the performance in character. More specifically, for example, interleaver length: L _turbo = 512 bits, N = 10, M = P = 53 (L _turbo / N ≦ P + 1), primitive root: g ₀ = 2 and mapping pattern: c It is created using the formula (1).

  As a result, the mapping pattern C is {1, 2, 4, 8, 16, 32, 11, 22, 44, 35, 17, 34, 15, 30, 7, 14, 28, 3, 6, as in the conventional case. 12, 24, 48, 43, 33, 13, 26, 52, 51, 49, 45, 37, 21, 42, 31, 9, 18, 36, 19, 38, 23, 46, 39, 25, 50, 47, 41, 29, 5, 10, 20, 40, 27}.

Further, in the PPI, the mapping pattern C is represented by (g ₀ × c (i−1)) mod P in units of rows, that is, 1, 2, 4, 8, 16, 32, 11, Bits are exchanged by shifting 22 and 44 at a time, and a j-row mapping pattern: C _j is generated. FIG. 9 is a diagram showing mapping patterns C _j rearranged by the above method. Note that j = 0 to N-1.

FIG. 10 is a diagram showing a data arrangement when data of interleaver length: L _turbo = 512 bits is mapped to the rearranged mapping pattern. Here, data {0-52} is stored in the first line, data {53-105} is stored in the second line, data {106-158} is stored in the third line, and data {159-211} is stored in the fourth line. Data {212-264} on the 5th line, data {265-317} on the 6th line, data {318-370} on the 7th line, data {371-423} on the 8th line, 9 Data {424 to 476} is mapped to the line, and data {477 to 529} is mapped to the 10th line.

  Finally, FIG. 11 is a diagram showing a final rearrangement pattern. Here, in accordance with a predetermined rule, the rows in the data array in FIG. 10 are exchanged to generate a final rearrangement pattern. In this embodiment, the order of the rows in the data array in FIG. 10 is reversed. In the PPI, the generated rearrangement pattern is read in units of columns, that is, vertically.

  FIG. 12 is a diagram in which the PPI of the present embodiment and the conventional PIL are quantitatively compared. The minimum distance between signal points (1, x) here means the minimum distance between signal points in adjacent rows in an N × M buffer, and the minimum distance between signal points (2, x). Is an N × M buffer that is read from the left column in the vertical order and connected serially to the next data in the array, or between the signal points of the next data This refers to the minimum distance, and is described in order up to the minimum distance between signal points with the 9th data or 9th data. Variance (or expressed as Normalized dispersion) is used as an index representing randomness, and 1 is the maximum (100% random).

  As described above, in this embodiment, by using the PPI of the present invention as an interleaver, the randomness can be improved while maintaining the same signal point distance as the conventional one, so that the error correction capability is improved. It becomes possible to make it. As a result, the demodulation characteristics on the receiving side can be greatly improved, so that an optimum transmission characteristic close to the Shannon limit, that is, an optimum BER characteristic can be obtained.

  In this embodiment, rearrangement was performed under the same conditions as in the prior art for performance comparison, but each parameter such as interleave length is arbitrary and can be changed as appropriate.

Embodiment 2. FIG.
FIGS. 13, 14, and 15 are diagrams showing processing of the second embodiment of the interleavers 32 and 33 used in the turbo encoder shown in FIG. Here, a process of randomly exchanging information bit sequences using interleavers 32 and 33 according to the present embodiment will be described. Since the configuration other than the interleaver is the same as that of the first embodiment, the same reference numerals are given and description thereof is omitted.

The PPI of this embodiment has the following three features.
(1) The rows and columns in the N (vertical axis: natural number) × M (horizontal axis: natural number) buffer are switched.
(2) The mapping pattern is shifted one column at a time (left) in units of rows.
(3) Avoid critical patterns by replacing rows.

Here, the operation of the PPI that is the interleaver of the present embodiment will be described. In the present embodiment, rearrangement is performed under the same conditions as the interleaver described in the prior art in order to evaluate the performance in character. More specifically, for example, interleaver length: L _turbo = 512 bits, N = 10, M = P = 53 (L _turbo / N ≦ P + 1), primitive root: g ₀ = 2 and mapping pattern: c It is created using the formula (1).

  As a result, the mapping pattern C is {1, 2, 4, 8, 16, 32, 11, 22, 44, 35, 17, 34, 15, 30, 7, 14, 28 as in the conventional and the first embodiment. 3, 6, 12, 24, 48, 43, 33, 13, 26, 52, 51, 49, 45, 37, 21, 42, 31, 9, 18, 36, 19, 38, 23, 46, 39 25, 50, 47, 41, 29, 5, 10, 20, 40, 27}.

In the PPI, the mapping pattern C, and row unit, the row unit performs bit permutation by shifting one column, j row of the mapping pattern: generates a C _j. FIG. 13 is a diagram showing mapping patterns C _j rearranged by the above method. Note that j = 0 to N-1.

FIG. 14 is a diagram showing a data arrangement when data of interleaver length: L _turbo = 512 bits is mapped to the rearranged mapping pattern. Here, data {0-52} is stored in the first line, data {53-105} is stored in the second line, data {106-158} is stored in the third line, and data {159-211} is stored in the fourth line. Data {212-264} on the 5th line, data {265-317} on the 6th line, data {318-370} on the 7th line, data {371-423} on the 8th line, 9 Data {424 to 476} is mapped to the line, and data {477 to 529} is mapped to the 10th line.

  Finally, FIG. 15 is a diagram illustrating a final rearrangement pattern. Here, according to a predetermined rule, the rows in the data array in FIG. 14 are exchanged to generate a final rearrangement pattern. In the present embodiment, the order of the rows in the data array in FIG. 14 is reversed. In the PPI, the generated rearrangement pattern is read in units of columns, that is, vertically.

  FIG. 16 is a diagram in which the PPI of the present embodiment and the conventional PIL are quantitatively compared. The minimum distance between signal points (1, x) here means the minimum distance between signal points in adjacent rows in an N × M buffer, and the minimum distance between signal points (2, x). Is an N × M buffer that is read from the left column in the vertical order and connected serially to the next data in the array, or between the signal points of the next data This refers to the minimum distance, and is described in order up to the minimum distance between signal points with the 9th data or 9th data. Variance (or expressed as Normalized dispersion) is used as an index representing randomness, and 1 is the maximum (100% random).

  As described above, in the present embodiment, by using the PPI of the present invention as interleaving, the distance between signal points can be greatly improved as compared with the prior art. It becomes possible to further improve the error correction capability. As a result, the demodulation characteristic on the receiving side can be further greatly improved, so that an optimum transmission characteristic close to the Shannon limit, that is, an optimum BER characteristic can be obtained.

Embodiment 3 FIG.
When a turbo encoder as shown in FIG. 5A is used, if the interleaver 32 and the interleaver 33 are the same, the distance between signal points is zero. Therefore, for example, when one output is affected by noise or the like, the other output is similarly affected. Therefore, in the present embodiment, an interleaver optimal for a turbo encoder having two information bit sequences as shown in FIG. 5A, that is, a sufficient distance between signal points between information bit sequences can be taken. Provide an interleaver that can.

  Hereinafter, the operation of the interleaver of the present embodiment will be described using the interleaver of the first embodiment. For example, in the present embodiment, rearrangement is performed so that the distance between signal points of two information bit sequences does not become zero. Specifically, for example, the information bit sequence U1 is arranged in the order of the first row, the second row, the third row,..., The tenth row of the N × M buffer in the interleaver 32, and at the same time, the information bit sequence. : U2 is arranged in the order of the fifth, sixth, seventh,..., Fourth line of the N × M buffer in the interleaver 33. Then, after the rearrangement process is completed as in the first embodiment, the respective data series arranged in the N × M buffers in both interleavers are read vertically in order from the first column of the first row.

  FIG. 17 is a diagram illustrating the distance between the signal points U1 and U2 when the distance between the information bit sequence U1 and the information bit sequence U2 is 5 rows as described above. In the present embodiment, not only when the distance between U1 and U2 is 5 lines, but also when the distance between U1 and U2 is 1 line to 9 lines by the same method as described above. Find all signal-to-signal distances. In this embodiment, rearrangement is performed using two interleavers having a distance between U1 and U2 from which optimum transmission characteristics can be obtained.

  FIG. 18 is a diagram illustrating optimum values of two interleavers when rearrangement is performed under the same conditions as in the first embodiment. Here, when the distance between the information bit sequence: U1 and the information bit sequence: U2 is 9 rows, optimum transmission characteristics can be obtained.

  As described above, in this embodiment, the optimum transmission path is obtained by obtaining all the distances between the information bit series: U1 and the information bit series: U2 and further obtaining all the inter-signal distances for each distance. Select an interleaver that provides the characteristics. As a result, an interleaver that is optimal for a turbo encoder having two information bit sequences as shown in FIG. 5A, that is, an interleaver that can sufficiently take the distance between signal points between information bit sequences is realized. can do.

  In the present embodiment, the first embodiment is used for convenience of explanation. However, the present invention is not limited to this, and the present invention may be applied to the interleaver of the second embodiment, and the other two information bits. The present invention may be applied to a turbo encoder having a sequence.

  In this embodiment, the two information bit sequences are arranged in different rows of the N × M buffer in the interleaver 32. However, the present invention is not limited to this. For example, the two information bit sequences are interleaved. 32 may be arranged at the same position in the N × M buffer in 32 and the read start position after rearrangement may be shifted.

  DESCRIPTION OF SYMBOLS 1 Turbo encoder, 11 1st decoder, 12, 16, 65, 66, 67, 68, 69, 75, 76, 77, 78 Adder, 13, 14, 32, 33 Interleaver, 15 2nd decoding 17 deinterleaver, 18 first determiner, 19 first R / S decoder, 20 second determiner, 21 second R / S decoder, 22 third determiner, 31 first recursion Systematic convolutional encoder, 34 second recursive systematic convolutional encoder, 41 multiplex / sync control, 42, 43 cyclic redundancy check (CRC), 44, 45 forward error collection (FEC), 46 Interleave, 47, 48 Rate converter, 49 Tone ordering, 50 Constellation encoder / gain scaling, 51 Reverse height Fourier transform unit (IFFT), 52 input parallel / serial buffer, 53 analog processing / digital-analog converter, 61, 62, 63, 64, 71, 72, 73, 74 delay unit, 141 analog processing / analog-digital converter, 142 Time Domain Equalizer (TEC), 143 Input Serial / Parallel Buffer, 144 Fast Fourier Transform (FFT), 145 Frequency Domain Equalizer (FEC), 146 Constellation Encoder / Gain Scaling, 147 Tone Ordering, 148, 149 Rate Converter, 150 Deinterleave, 151,152 Forward error collection, 153,154 Cyclic redundancy check (CRC , 155 multiplex / sync control.

Claims (4)

In an interleaving apparatus that performs interleaving processing on a first data series in which a plurality of data is arranged in a first arrangement order, and generates a second data series in which the plurality of data is arranged in a second arrangement order ,
A sequence composed of a first predetermined number of elements corresponding to the number of the plurality of data, wherein the elements are expressed in an arrangement order in the first arrangement order and arranged in the second arrangement order. The series is a mapping series,
A third predetermined number set ((first predetermined number) = (second predetermined number) × (third predetermined number)) is obtained by dividing the mapping sequence into a second predetermined number of elements. Each series is a first submapping series,
A sequence composed of a third predetermined number of elements arranged in the same arrangement order in each of the first sub-mapping sequences, and each of the second predetermined number of sets as a second sub-mapping sequence,
Of the second predetermined number of sets of second sub-mapping sequences, a second sub-mapping sequence including a specific element is a basic sequence.
Of the second predetermined number of sets of second sub-mapping sequences, at least one second sub-mapping sequence excluding the basic sequence is the basic sequence expressed in the arrangement order in the first arrangement order. The interleaving apparatus is equivalent to a sequence obtained by adding a predetermined number to each of the elements and cyclically shifting the resulting sequence.   Of the second predetermined number of second submapping sequences, each second submapping sequence excluding the basic sequence has a predetermined number different from each other for each second submapping sequence. It is equivalent to any one of the respective sequences obtained by adding each element of the basic sequence expressed by the arrangement order in the arrangement order of each of them, and cyclically shifting each of the obtained sequences. The interleaving device according to claim 1.   The interleaving apparatus according to claim 1, wherein the specific element is a first arrangement order in the first arrangement order. In an interleaving apparatus that performs interleaving processing on a first data series in which a plurality of data is arranged in a first arrangement order, and generates a second data series in which the plurality of data is arranged in a second arrangement order ,
A sequence of a first predetermined number of elements according to the number of the plurality of pieces of data expressed in the arrangement order in the first arrangement order and arranged in the second arrangement order is a mapping series.
The interleaving apparatus is characterized in that the mapping sequence is equivalent to a sequence obtained based on a predetermined basic sequence including an element of the first arrangement order in the first arrangement order.

Images