An adaptive soft-decision decoding method for Zipper codes based on Chase-Pyndiah algorithm

Through the Chase-Pyndiah algorithm that adapts to adjust the number of unreliable positions, the balance between the coding complexity and performance of Zipper code soft decision is solved, reducing the algorithm complexity and maintaining performance.

CN114499547BActive Publication Date: 2024-11-05XIDIAN UNIV
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202210005063.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-01-04
Publication Date
2024-11-05
Estimated Expiration
2042-01-04

AI Technical Summary

Technical Problem

In the prior art, the soft decision decoding of Zipper code based on Chase-Pyndiah algorithm is difficult to achieve a balance between complexity and performance. Especially when the channel conditions are good, the generation of a large number of test sequences increases the complexity, and for channel conditions In good cases, there is no need for too many test sequences.

Method used

By adaptively adjusting the number of unreliable positions, dynamically adjusting the number of test sequences in the Chase-Pyndiah algorithm, and decoding using channel information and soft information, the compromise between complexity and performance of Zipper code soft decoding is achieved.

Benefits of technology

It effectively reduces the complexity of the algorithm, reduces the number of test sequences, maintains performance, and achieves a balance between complexity and performance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114499547B_ABST
    Figure CN114499547B_ABST
Patent Text Reader

Abstract

The present invention discloses a Zipper code adaptive soft decision decoding method based on Chase-Pyndiah algorithm, including: initializing the decoding information of the real block and the imaginary block of the Zipper code in the current decoding window; determining the number of unreliable positions of each row of component code in the current decoding window, and decoding; updating the external information value of each bit and exchanging information between the real block and the imaginary block according to the mapping relationship to complete one iteration; obtaining the prior value of the next iteration of the current decoding window according to the external information value of each bit after the update, and performing the next iteration; when the preset maximum number of iterations is reached, the decoder outputs the decoding result of the target codeword block; sliding the decoding window to include the next codeword block to decode the next codeword block. The present invention makes full use of channel information and soft information, dynamically and adaptively adjusts the number of unreliable positions, achieves a compromise between the complexity and performance of Zipper code soft decoding, and effectively reduces the complexity.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the field of communication technology, and in particular relates to a Zipper code adaptive soft decision decoding method based on a Chase-Pyndiah algorithm. Background Art

[0002] In 2000, forward error correction (FEC) in optical transport networks (OTN) was first standardized in ITU-T Recommendation G.975, which specifies (255,239) Reed-Solomon codes operating at a throughput of 2.5Gb / s. Since then, more complex FEC schemes for OTN have been studied and implemented to accommodate higher data rates. Since OTN handles very high data throughput, the FEC used must have lower complexity. Spatially coupled codes are a type of FEC suitable for OTN networks, including spatially coupled LDPC (Low Density Parity Check Code), Braided Block codes, and Staircase codes.

[0003] Zipper code is a new spatial coupling code framework proposed by AYSukmadji et al. in 2019. Due to complexity considerations, FEC in OTN generally uses hard decision decoding, but when the complexity requirement is not too high and the bit error rate requirement is high, hard decision decoding cannot meet the requirements. Based on this, the prior art proposes a soft decision decoding of Zipper code based on Chase-Pyndiah, but the algorithm will generate a large number of test sequences during the decoding process, some of which do not contribute much to decoding, and when the channel conditions are good, not too many test sequences are needed, which adds unnecessary complexity. Summary of the invention

[0004] In order to solve the above problems existing in the prior art, the present invention provides a Zipper code adaptive soft decision decoding method based on the Chase-Pyndiah algorithm. The technical problem to be solved by the present invention is achieved by the following technical solutions:

[0005] The present invention provides a Zipper code adaptive soft decision decoding method based on Chase-Pyndiah algorithm, comprising:

[0006] S1: Initialize the decoding information of the real block and imaginary block of the Zipper code in the current decoding window;

[0007] S2: determining the number of unreliable positions of each row of component codes in the current decoding window according to the channel conditions, and using the unreliable position number for decoding;

[0008] S3: updating the external information value of each bit according to the decoding result of the current decoding window and performing information exchange between the real part block and the imaginary part block according to the mapping relationship to complete one iteration;

[0009] S4: obtaining a priori value for the next iteration of the current decoding window according to the updated external information value of each bit, and performing the next iteration;

[0010] S5: When the preset maximum number of iterations is reached, the decoder outputs the decoding result of the target codeword block;

[0011] S6: Slide the decoding window to include the next codeword block to decode the next codeword block.

[0012] In one embodiment of the present invention, the S1 includes:

[0013] If the current decoding window is the first decoding window to start decoding, the prior values ​​of the real part blocks of the Zipper code in the first decoding window are initialized to 0, and the channel information of the real part blocks is initialized to its received value; the prior value of the first imaginary part block of the Zipper code in the first decoding window is initialized to 0, and the channel information of the first imaginary part block is initialized to -1.

[0014] In one embodiment of the present invention, the S1 further includes:

[0015] If the current decoding window is a decoding window after the first decoding window, the a priori value of the most recently received real part block in the current decoding window is initialized to 0, the a priori value of the most recently received previous real part block is initialized to the external information of the real part bit corresponding to the previous real part block at the last iteration of the previous decoding window, the a priori values ​​of the remaining real part blocks in the current decoding window are initialized to the external information of the corresponding imaginary part bits obtained according to the mapping relationship at the last iteration of the previous decoding window, and the channel information of all real part blocks is initialized to the received value;

[0016] The a priori value of the first imaginary part block in the current decoding window is initialized to the external information of the imaginary part bit corresponding to the first imaginary part block in the last iteration of the previous decoding window, and the channel information of all imaginary part blocks in the current decoding window are the corresponding real part bit reception values ​​obtained according to the mapping relationship.

[0017] In one embodiment of the present invention, the S2 includes:

[0018] S21: Obtain soft input information of the i-th row of codewords in the current decoding window:

[0019]

[0020] in, α is the scaling factor, represents the channel information of the bit in the i-th row and j-th column, represents the prior value of the bit in the i-th row and j-th column at the I-th iteration, where I represents the number of iterations;

[0021] S22: determining the number of unreliable positions of each row of component codes in the current decoding window according to the soft input information;

[0022] S23: Decoding is performed using the Chase decoder to obtain a decoding result of the current decoding window.

[0023] In one embodiment of the present invention, the S22 includes:

[0024] S221: Count the average value of all the bit soft input information in the i-th row component code whose absolute value is less than 1

[0025] S222: Obtaining soft input information in the i-th row component code The absolute value of is less than The number of bits q, where ρ is the dynamic adjustment factor;

[0026] S223: Obtain the number of unreliable positions of the component code in the i-th row, expressed as:

[0027]

[0028] In one embodiment of the present invention, the S23 includes:

[0029] According to the unreliable position number p and the input soft information of the current row component code, the p most unreliable bits in the current row component code are selected, and the p most unreliable bits are flipped to generate 2 p test sequence, and the 2 p The test sequence is decoded and the decoding result of the component code of the current row is obtained.

[0030] In one embodiment of the present invention, S3 includes:

[0031] S31: Update the external information value of each bit using the decoding result of each row of component codes:

[0032]

[0033] Where j = 0, 1...n-1, is the modulation form of the final output codeword of the decoder, is the modulation form of other codewords in the candidate codeword list, β is the reliability factor, is the final output codeword of the decoder The jth bit of

[0034] S32: after decoding of each row of component codes is completed, the updated external information value of the real part is returned to the imaginary part according to the mapping relationship in the current decoding window as a priori value of the imaginary part;

[0035] S33: repeating steps S31 and S32, decoding each row of component code in the current decoding window in turn, and completing one iteration of the current decoding window.

[0036] In one embodiment of the present invention, the S4 includes:

[0037] S41: After one iteration is completed, the updated external information of all imaginary part blocks except the first imaginary part in the current decoding window is returned to the corresponding real part according to the mapping relationship as the prior value of the real part in the next iteration; the external information of the first imaginary part block and the last real part block are used as the respective prior values ​​in the next iteration;

[0038] S42: Iteration number I=I+1, repeat S3, and decode from the first row to the last row in the current decoding window.

[0039] Another aspect of the present invention provides a storage medium storing a computer program for executing the steps of the Zipper code adaptive soft decision decoding method based on the Chase-Pyndiah algorithm described in any one of the above embodiments.

[0040] Another aspect of the present invention provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor calls the computer program in the memory, the steps of the Zipper code adaptive soft decision decoding method based on the Chase-Pyndiah algorithm as described in any one of the above embodiments are implemented.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] The invention discloses a Zipper code adaptive soft decision decoding method based on a Chase-Pyndiah algorithm, sets the number of unreliable positions to be adaptively variable, makes full use of channel information and soft information, dynamically and adaptively adjusts the number of unreliable positions, and achieves a compromise between the complexity and performance of the Zipper code soft decoding. When the number of unreliable positions is reduced by one, the test sequence of the Chase-Pyndiah algorithm is reduced by half, and the complexity is also significantly reduced accordingly, thereby effectively reducing the algorithm complexity when the performance difference is not much.

[0043] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a structural block diagram of a Zipper encoding and decoding simulation system;

[0045] Figure 2 It is a schematic diagram of the structure of a Zipper code;

[0046] Figure 3 It is a flowchart of a Zipper code adaptive soft decoding method based on the Chase-Pyndiah algorithm provided by an embodiment of the present invention;

[0047] Figure 4 It is a flow chart of a Zipper code adaptive soft decoding method based on the Chase-Pyndiah algorithm provided by an embodiment of the present invention;

[0048] Figure 5 Schematic diagram of a Zipper code sliding window decoding provided by an embodiment of the present invention;

[0049] Figure 6 It is a principle block diagram of a Chase-Pyndiah algorithm provided by an embodiment of the present invention;

[0050] Figure 7 is a flow chart of an algorithm for determining the number of unreliable positions provided by an embodiment of the present invention;

[0051] Figure 8 1 is a schematic diagram of a mapping relationship of a Zipper code provided in an embodiment of the present invention;

[0052] Fig. 9 These are simulation results obtained using the Zipper code adaptive soft decision decoding method and the decoding method with a fixed number of unreliable positions according to the embodiment of the present invention;

[0053] Fig.10 This is a simulation result diagram showing how the signal-to-noise ratio changes with the number of unreliable positions under different dynamic factors. DETAILED DESCRIPTION

[0054] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of a Zipper code adaptive soft decision decoding method based on the Chase-Pyndiah algorithm proposed in the present invention in combination with the accompanying drawings and specific implementation methods.

[0055] The above and other technical contents, features and effects of the present invention are clearly presented in the following detailed description of the specific implementation modes in conjunction with the accompanying drawings. Through the description of the specific implementation modes, the technical means and effects adopted by the present invention to achieve the predetermined purpose can be more deeply and specifically understood. However, the attached drawings are only for reference and explanation purposes and are not used to limit the technical solutions of the present invention.

[0056] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants are intended to cover non-exclusive inclusion, so that an article or device including a series of elements includes not only those elements, but also other elements that are not explicitly listed. In the absence of more restrictions, the elements defined by the statement "including one..." do not exclude the existence of other identical elements in the article or device including the elements.

[0057] First, let's introduce the concepts and encoding process of Zipper code. Figure 1 , Figure 1The invention discloses a structural block diagram of a Zipper coding simulation system, which comprises a pseudo-random sequence generation module, a Zipper code encoding module, a BPSK modulation module, an AWGN channel module, a BPSK demodulation module, a Zipper code decoding module and a statistical error module, wherein the pseudo-random sequence generation module is used to generate source bits through a random function, and send the generated source bits to the Zipper code encoding module; the Zipper code encoding module is used to perform Zipper code encoding on the input source bits, and send the bit sequence obtained after encoding to the BPSK modulation module; the BPSK modulation module is used to perform binary phase shift keying modulation on the bit sequence sent by the Zipper code encoding module, so that "0" becomes "-1" after modulation, and "1" becomes "+1" after modulation, and the modulated bit sequence is sent to the BPSK modulation module. The sequence is sent to the AWGN channel module; the AWGN channel module is used to add additive white Gaussian noise to the modulated sequence, and then send it to the BPSK demodulation module; the BPSK demodulation module is used to demap the bit sequence from the AWGN channel module, obtain the log-likelihood ratio information and the bit sequence after hard decision, and then send the obtained log-likelihood ratio information, the bit sequence after hard decision and the received sequence to the Zipper code decoding module; the Zipper code decoding module is used to perform soft decoding based on the log-likelihood ratio information sent by the BPSK demodulation module, the bit sequence after hard decision and the received sequence, obtain the final decision sequence and send the final decision sequence to the statistical error module; the statistical error module is used to compare the final decision sequence obtained by decoding with the generated pseudo-random sequence, count the number of error bits, and obtain the bit error rate.

[0058] For further information, see Figure 2 , Figure 2 This is a schematic diagram of the structure of a Zipper code. The left half of the figure is C A is the imaginary part, the right half C B is the real part, which includes information bits and check bits. Only the real part can be transmitted in the channel, and the imaginary part is not sent in the channel. Each time a codeword block is sent, the size of the codeword block is recorded as L, that is, a codeword block includes L row component codes of the Zipper code. Figure 2 Each row in is a legal codeword of the component code. For the i-th row, the imaginary part of the row is m i , the total code length is recorded as n i , the real part is n i -m i . If there exists a positive integer v such that m i+v =m i , then v is the period of the Zipper code.

[0059] The encoding process of the Zipper code is as follows: After the encoding is initialized, the information sequence is placed in the corresponding position of the real part, and is sent to the component encoder row by row for encoding to obtain the check bit. The check bit is placed in the corresponding position of the real part, and then the real part of the row is returned to the imaginary part according to the mapping relationship. At this point, the encoding of the row is completed, and the above process is continued for the next row until the encoding is completed.

[0060] After encoding of each L rows of component codes, the real part bits are BPSK modulated row by row and sent into the AWGN channel. The receiving end puts the received bits into the real part position row by row. Each L rows is a codeword block and is labeled. For example, the first L rows are marked as B0, the second L rows are marked as B1, and so on. The imaginary part is then obtained through the mapping relationship (the channel reception value of the first imaginary block A0 is initialized to -1). The mapping relationship from the imaginary part to the real part is φ, that is, the real part bit corresponding to the imaginary part bit in the i-th row and j-th column is φ(i,j); the mapping relationship from the real part to the imaginary part is φ -1 , that is, the imaginary bit corresponding to the real bit of the i-th row and j-th column is φ -1 (i,j).

[0061] r i =(r i,0 ,r i,1 ,...r i,j ,...,r i,n ) represents the received value of the component code channel in the i-th row, represents the channel information of the i-th row component code. Similarly, and They respectively represent the prior value and extrinsic information value of the i-th row component code at the I-th iteration. For vector The jth corresponding value of represents the channel information of the i-th row and j-th column bit. Similarly, The vectors are The j-th corresponding value of , which respectively represents the prior value and extrinsic information value of the i-th row and j-th column bit at the I-th iteration.

[0062] On this basis, this embodiment provides a Zipper code dynamic soft decoding method based on the Chase-Pyndiah algorithm, such as Figure 3 and Figure 4 , the method comprises the following steps:

[0063] S1: Initialize the decoding information of the real block and the imaginary block of the Zipper code in the current decoding window.

[0064] If the current decoding window is the first decoding window to start decoding, the prior values ​​of the real part blocks of the Zipper code in the first decoding window are initialized to 0, and the channel information of the real part blocks is initialized to its received value; the prior value of the first imaginary part block of the Zipper code in the first decoding window is initialized to 0, and the channel information of the first imaginary part block is initialized to -1.

[0065] If the current decoding window is a decoding window after the first decoding window, the a priori value of the most recently received real part block in the current decoding window is initialized to 0, the a priori value of the previous real part block is initialized to the external information of the real part bit corresponding to the previous real part block at the last iteration of the previous decoding window, the a priori values ​​of the remaining real part blocks in the current decoding window are initialized to the external information of the corresponding imaginary part bits obtained according to the mapping relationship at the last iteration of the previous decoding window, and the channel information of all real part blocks is initialized to the received value;

[0066] The a priori value of the first imaginary part block in the current decoding window is initialized to the external information of the imaginary part bit corresponding to the first imaginary part block in the last iteration of the previous window, and the channel information of all imaginary part blocks in the current decoding window are the received values ​​of the corresponding real part bits obtained according to the mapping relationship.

[0067] Specifically, assuming that the code length of the Zipper code is n, the size of the imaginary part is m, the size of the codeword block is L, and the decoding window size is w, decoding begins after receiving w codeword blocks. As mentioned above, the Zipper code includes a real part and an imaginary part, so the codeword block also includes a real part block and an imaginary part block. Before decoding begins, the decoding information of the real part block and the imaginary part block is first initialized.

[0068] Specifically, if the current decoding window is the first decoding window to start decoding, the w real part blocks B0, B1, ..., B w-1 The prior values ​​of are initialized to 0, w real part blocks B0, B1, ...., B w-1 The channel information of the first imaginary block A0 of the Zipper code in the first decoding window is initialized to 0, and the channel information of the first imaginary block A0 is initialized to -1. The above initialization process is expressed by the following formula:

[0069] (a) For the real part j∈[m,n-1]

[0070]

[0071] Where L represents the codeword block size, w represents the decoding window size, represents the channel information of the bit in the i-th row and j-th column, represents the initial prior value of the bit in the i-th row and j-th column, r i,j Represents the channel received value of the bit in the i-th row and j-th column.

[0072] (b) For the imaginary part j∈[0,m-1]

[0073]

[0074] See also Figure 5 , Figure 5 1 is a schematic diagram of a Zipper code sliding window decoding provided by an embodiment of the present invention, wherein a large dashed line frame represents a current window, a small dashed line frame represents a target block at a current moment in the current window, and decoding begins after receiving w codeword blocks. Figure 5 The window size is 3, that is, decoding starts after receiving w=3 codeword blocks. In this embodiment, before decoding starts, the first decoding window is initialized. For the real part, the prior values ​​of the three real blocks B0, B1, and B2 in the current window are initialized to 0, and the channel information of the three real blocks B0, B1, and B2 are initialized to their received values; for the imaginary part, the prior value of the first imaginary block A0 is initialized to 0, and the channel information of the first imaginary block A0 is initialized to -1. The window size of this embodiment is w=3, and the initialization process is expressed by the following formula:

[0075] (a) For the real part j∈[m,n-1]

[0076]

[0077] (b) For the imaginary part j∈[0,m-1]

[0078]

[0079] Furthermore, if the current decoding window is a decoding window after the first decoding window, that is, the current decoding window is not the first decoding window, if the target block of the current decoding window is B h , then for the real part, the most recently received real part block B in the current decoding window h+w-1 The prior value of is initialized to 0, the real part of block B h+w-1 The previous real part block B h+w-2 The a priori value of is initialized to the external information of the real part bit corresponding to the block at the last iteration of the previous decoding window, and the a priori values ​​of the remaining real part blocks in the current decoding window are initialized to the external information of the corresponding imaginary part bits obtained according to the mapping relationship at the last iteration of the previous decoding window. The channel information of all real part blocks is initialized to the received value; for the imaginary part, the first codeword block A of the current decoding window is initialized to h The prior value of the codeword block A is hThe external information corresponding to the imaginary bit at the last iteration of the previous window, and the imaginary channel information of all codeword blocks in the current decoding window are the received values ​​of the corresponding real bits obtained according to the mapping relationship. It can be expressed as follows:

[0080] (a) For the real part j∈[m,n-1]:

[0081]

[0082] (b) For the imaginary part j∈[0,m-1]:

[0083]

[0084] Specifically, Figure 5 As shown, for the case where the window size is w=3, if the target codeword block of the current decoding window is B h , then for the real part, the most recently received real part block B h+2 The real part of the prior value is initialized to 0, and the previous real part block B h+1 The real part prior value of the real part block B h+1 In the previous window (i.e. Figure 5 The h-1 window in the last iteration corresponds to the external information of the real part bit, the target codeword block B h The prior value of is the previous window (i.e. Figure 5 h-1 window in the last iteration) According to the external information of the corresponding imaginary bit obtained by the mapping relationship, all real block B h ,B h+1 ,B h+2 The channel information is the received value; for the imaginary part, the first imaginary block A in the current decoding window h The imaginary part of the prior value is the previous window A h External information at the last iteration, all A in the window h ,A h+1 ,A h+2 The channel information of the imaginary block is the corresponding real bit received value obtained according to the mapping relationship. At this time, it is expressed as follows:

[0085] (a) For the real part:

[0086]

[0087] (b) For the imaginary part:

[0088]

[0089] S2: Determine the number of unreliable positions of each row of component code in the current decoding window according to the channel conditions, and perform decoding based on the unreliable positions.

[0090] Specifically, decoding starts after the decoding window is initialized, starting from the first row of component codes in the current decoding window. This embodiment uses the Chase-Pyndiah algorithm to decode the component codes in the decoding window. The Chase-Pyndiah algorithm is an improved algorithm based on the Chase algorithm. The Chase algorithm generates different error patterns according to the reliability of each bit. Each error pattern corresponds to a test sequence. These test sequences are then decoded using a hard decision decoder. After decoding, the codewords that meet the verification are output to form a list of candidate codewords. The codewords in the list are then compared with the received sequence, and a candidate codeword with the smallest Euclidean distance to the received sequence is selected as the final output. The Chase-Pyndiah algorithm adds the calculation of output external information on the basis of the Chase algorithm. Its basic principles are as follows: Figure 6 shown.

[0091] In this embodiment, step S2 includes:

[0092] S21: Obtain soft input information of the i-th row component code in the current decoding window:

[0093] At the Ith iteration, is the soft input information of the i-th row codeword, in, represents the channel information of the bit in the i-th row and j-th column, represents the prior value of the bit in the i-th row and j-th column at the I-th iteration, where I represents the number of iterations.

[0094] S22: Determine the number of unreliable positions of each row of component codes in the current decoding window according to the soft input information.

[0095] For each row of component code, the average value of the bit soft input information whose absolute value is less than 1 in all bit soft input information of the component code is calculated. Specifically, select the bit soft input information whose absolute value of all bit soft input information in the component code is less than 1, and calculate the average value of these bit soft input information, which is

[0096] Assuming a dynamic adjustment factor ρ, the soft input information in the component code is counted The absolute value of is less than The number of bits q is determined according to the value of q. If q>4, in order to avoid high decoding complexity, set p=4; if q<2, in order to reduce performance loss, set p=2; for other cases, set p=q, which is expressed as follows:

[0097]

[0098] Preferably, the dynamic adjustment factor ρ is set to a value of 0.6-1.0. The larger the dynamic adjustment factor is, the larger the number of unreliable positions is, and the higher the algorithm complexity is.

[0099] It should be noted that the additional complexity of the Zipper code soft decision decoding algorithm based on Chase-Pyndiah is exponentially related to the number of unreliable positions p, and linearly related to the code length n of the component code. For a specific component code, the code length n is a fixed value. If p is set to a fixed value, such as setting p = 5, 2 5 = 32 test sequences, but in the decoding process, some test sequences do not contribute much to the decoding result, and when the channel conditions are good, too many test sequences are not needed. If the number of unreliable positions is a fixed value, it will add a lot of complexity. Therefore, this embodiment sets the number of unreliable positions to be adaptively variable, achieving a compromise between the complexity and performance of Zipper code soft decoding, and reducing the complexity when the performance difference is not too large.

[0100] S23: Decoding is performed using the Chase decoder to obtain a decoding result of the current decoding window.

[0101] Specifically, according to the number of unreliable positions p and the input soft information of the current row component code, the p most unreliable bits in the current row component code are selected, and the p most unreliable bits are flipped to generate 2 p Test sequences (a bit has two conditions: flip and not flip, so a total of 2 p Flip: Assume the bit is 0, and it will be 1 after flipping; Do not flip: Assume the bit is 0, and it will still be 0 if it is not flipped). p The test sequence is decoded and the decoding result of the component code of the current row is obtained.

[0102] S3: Update the external information value of each bit according to the decoding result of the current decoding window and exchange information between the real part block and the imaginary part block according to the mapping relationship to complete one iteration.

[0103] Specifically, step S3 of this embodiment includes:

[0104] S31: Update the external information value of each bit using the decoding result of each row of component codes:

[0105]

[0106] Where j = 0, 1...n-1, is the modulation form of the final output codeword of the decoder, is the modulation form of other codewords in the candidate codeword list, which may not exist; β is a reliability factor, which can be optimized according to the number of iterations and channel conditions. In this embodiment, β=1 is selected; The final output codeword for the decoder The jth bit of

[0107] S32: After decoding of each row of component codes is completed, the external information value updated by the real part is returned to the imaginary part according to the mapping relationship in the current decoding window as the prior value of the imaginary part. The information exchange process is as follows:

[0108]

[0109] S33: Repeat steps S31 and S32 to decode each row of component codes in the current decoding window in turn to complete one iteration of the current decoding window, that is, decoding from the first row to the last row in the current decoding window is one iteration.

[0110] S4: Obtain the priori value of the next iteration of the current decoding window according to the updated external information value of each bit, and perform the next iteration according to the process of step S3.

[0111] Specifically, after one iteration is completed, the updated external information of all imaginary blocks except the first imaginary part in the current decoding window is returned to the corresponding real part according to the mapping relationship, which is used as the prior value of the real part in the next iteration; the external information of the first imaginary block and the last real block are used as their respective prior values ​​in the next iteration; then, the number of iterations I=I+1, S3 is repeated, and decoding is started from the first row of the current decoding window to the last row.

[0112] Specifically, after one iteration, the last w-1 imaginary blocks A in the current decoding window are h+1 ,A h+2 ,...A h+w-1 The updated external information returns the real part according to the mapping relationship as the prior value of the real part in the next iteration; the first imaginary block A h and the last real block B h+w-1 The external information is used as the prior value of each in the next iteration, and the specific update is as follows:

[0113] (a) For the real part j∈[m,n-1]:

[0114]

[0115] (b) For the imaginary part j∈[0,m-1]:

[0116]

[0117] exist Figure 5That is: for the two imaginary blocks A in the current decoding window h+1 ,A h+2 The updated external information returns the real part according to the mapping relationship as the prior value of the corresponding real part in the next iteration; the first imaginary block A h and the last real block B h+2 The external information is used as the next iteration A h and B h+2 The prior value of

[0118] Then, the number of iterations I=I+1, and then steps S2-S3 are repeated, starting from the first row of the current decoding window to the last row.

[0119] S5: When the preset maximum number of iterations I is reached max Finally, the decoder outputs the decoding result of the target codeword block.

[0120] Specifically, repeat S2-S4 until the preset maximum number of iterations I is reached. max After that, the decoder outputs the target codeword block B h The decoding result of .

[0121] S6: Slide the decoding window to include the next codeword block to decode the next codeword block.

[0122] Specifically, the target codeword block B in the current decoding window h After decoding is completed, at the next moment, a new information block B is received h+w (like Figure 5 B h+3 As shown), the decoding window slides "down" by one block, with B h+1 The previous process is repeated as the target codeword block to start decoding the new codeword block until all codeword blocks are decoded.

[0123] The effect of the Zipper code adaptive soft decision decoding method according to the embodiment of the present invention is further described below in combination with simulation experiments.

[0124] (1) Simulation conditions

[0125] The simulation experiment of the embodiment of the present invention is carried out under Visual Studio 2019 software, and the Zipper code length n=62, the real part and the imaginary part size are both m=31, the period v=1, the block size is 31, the window size w=7, and the maximum number of decoding iterations I max =7, the mapping relationship is as follows Figure 8 As shown (it should be noted that m=31 in this embodiment, while 5 is shown in the figure, and the rest remain unchanged), the component code uses the (63, 57) BCH code that corrects one error.

[0126] (2) Simulation content and result analysis

[0127] Under the above conditions, the method implemented by the present invention is compared with the existing soft decision decoding method. The results are as follows Fig. 9 and Fig.10 shown. Fig. 9 : are simulation results obtained by using the Zipper code adaptive soft decision decoding method and the decoding method with a fixed number of unreliable positions according to the embodiment of the present invention, respectively, wherein the horizontal axis represents the signal-to-noise ratio E b N0, in decibels dB, the vertical axis represents the bit error rate BER; Fig.10 This is a simulation result diagram of the signal-to-noise ratio changing with the number of unreliable positions under different dynamic factors, where the horizontal axis represents the signal-to-noise ratio E b N0, in decibels dB, the vertical axis represents the number of unreliable positions. Fig. 9 and Fig.10 It can be seen that the bit error rate curve with dynamic adjustment factor ρ=1.0 is not much different from the bit error rate curve with unreliable position number 4, but the average number of unreliable bits of the curve with dynamic adjustment factor ρ=1.0 is less than 4 and less than 3.7, and as the signal-to-noise ratio increases, its average number of unreliable positions approaches 2, thus reducing the complexity to a great extent.

[0128] This embodiment is based on the Zipper code adaptive soft decision decoding method of the Chase-Pyndiah algorithm, sets the number of unreliable positions to be adaptively variable, makes full use of channel information and soft information, dynamically and adaptively adjusts the number of unreliable positions, and achieves a compromise between the complexity and performance of the Zipper code soft decoding. When the number of unreliable positions is reduced by one, the test sequence of the Chase-Pyndiah algorithm is reduced by half, and the complexity is significantly reduced accordingly, thereby effectively reducing the algorithm complexity when the performance difference is not much.

[0129] Another embodiment of the present invention provides a storage medium, wherein a computer program is stored in the storage medium, and the computer program is used to execute the steps of the Zipper code adaptive soft decision decoding method based on the Chase-Pyndiah algorithm described in the above embodiment. Another aspect of the present invention provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor calls the computer program in the memory, the steps of the Zipper code adaptive soft decision decoding method based on the Chase-Pyndiah algorithm described in the above embodiment are implemented. Specifically, the above-mentioned integrated module implemented in the form of a software function module can be stored in a computer-readable storage medium. The above-mentioned software function module is stored in a storage medium, including several instructions for enabling an electronic device (which can be a personal computer, a server, or a network device, etc.) or a processor (processor) to execute some steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk.

[0130] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the protection scope of the present invention.

Claims

1. A Zipper code adaptive soft decision decoding method based on Chase-Pyndiah algorithm, characterized in that: include: S1: Initialize the decoding information of the real block and imaginary block of the Zipper code in the current decoding window; S2: determining the number of unreliable positions of each row of component codes in the current decoding window according to the channel conditions, and using the unreliable position number for decoding; S3: updating the external information value of each bit according to the decoding result of the current decoding window and performing information exchange between the real part block and the imaginary part block according to the mapping relationship to complete one iteration; S4: obtaining a priori value for the next iteration of the current decoding window according to the updated external information value of each bit, and performing the next iteration; S5: When the preset maximum number of iterations is reached, the decoder outputs the decoding result of the target codeword block; S6: sliding the decoding window to include the next codeword block to decode the next codeword block, The S2 includes: S21: Obtain soft input information of the i-th row of codewords in the current decoding window: in, α is the scaling factor, represents the channel information of the bit in the i-th row and j-th column, represents the prior value of the bit in the i-th row and j-th column at the I-th iteration, where I represents the number of iterations; S22: determining the number of unreliable positions of each row of component codes in the current decoding window according to the soft input information; S23: Decode using the Chase decoder to obtain the decoding result of the current decoding window. The S22 includes: S221: Count the average value of all the bit soft input information whose absolute value is less than 1 in the component code of the i-th row S222: Obtaining soft input information in the i-th row component code The absolute value of is less than The number of bits q, where ρ is the dynamic adjustment factor; S223: Obtain the number of unreliable positions of the component code in the i-th row, expressed as: The S23 includes: According to the unreliable position number p and the input soft information of the current row component code, the p most unreliable bits in the current row component code are selected, and the p most unreliable bits are flipped to generate 2 p test sequence, and the 2 p The decoding result of the component code of the current row is The S3 includes: S31: Update the external information value of each bit using the decoding result of each row of component codes: Where j = 0, 1...n-1, is the modulation form of the final output codeword of the decoder, is the modulation form of other codewords in the candidate codeword list, β is the reliability factor, is the final output codeword of the decoder The jth bit of S32: after decoding of each row of component codes is completed, the updated external information value of the real part is returned to the imaginary part according to the mapping relationship in the current decoding window as a priori value of the imaginary part; S33: repeating steps S31 and S32, decoding each row of component code in the current decoding window in turn, and completing one iteration of the current decoding window.

2. The Zipper code adaptive soft decision decoding method based on the Chase-Pyndiah algorithm according to claim 1, characterized in that: The S1 includes: If the current decoding window is the first decoding window to start decoding, the prior values ​​of the real part blocks of the Zipper code in the first decoding window are initialized to 0, and the channel information of the real part blocks is initialized to its received value; the prior value of the first imaginary part block of the Zipper code in the first decoding window is initialized to 0, and the channel information of the first imaginary part block is initialized to -1.

3. The Zipper code adaptive soft decision decoding method based on the Chase-Pyndiah algorithm according to claim 2, characterized in that: The S1 further comprises: If the current decoding window is a decoding window after the first decoding window, the a priori value of the most recently received real part block in the current decoding window is initialized to 0, the a priori value of the most recently received previous real part block is initialized to the external information of the real part bit corresponding to the previous real part block at the last iteration of the previous decoding window, the a priori values ​​of the remaining real part blocks in the current decoding window are initialized to the external information of the corresponding imaginary part bits obtained according to the mapping relationship at the last iteration of the previous decoding window, and the channel information of all real part blocks is initialized to the received value; The a priori value of the first imaginary part block in the current decoding window is initialized to the external information of the imaginary part bit corresponding to the first imaginary part block in the last iteration of the previous decoding window, and the channel information of all imaginary part blocks in the current decoding window are the corresponding real part bit reception values ​​obtained according to the mapping relationship.

4. The Zipper code adaptive soft decision decoding method based on the Chase-Pyndiah algorithm according to claim 3, characterized in that: The S4 includes: S41: After one iteration is completed, the updated external information of all imaginary part blocks except the first imaginary part in the current decoding window is returned to the corresponding real part according to the mapping relationship as the prior value of the real part in the next iteration; the external information of the first imaginary part block and the last real part block are used as the respective prior values ​​in the next iteration; S42: Iteration number I=I+1, repeat S3, and decode from the first row to the last row in the current decoding window.

5. A storage medium, characterized in that: The storage medium stores a computer program, which is used to execute the steps of the Zipper code adaptive soft decision decoding method based on the Chase-Pyndiah algorithm as described in any one of claims 1 to 4.

6. An electronic device, characterized in that: The invention comprises a memory and a processor, wherein a computer program is stored in the memory, and when the processor calls the computer program in the memory, the steps of the Zipper code adaptive soft decision decoding method based on the Chase-Pyndiah algorithm as claimed in any one of claims 1 to 4 are implemented.

Citation Information

Patent Citations

  • Systems and methods for advanced iterative decoding and channel estimation of concatenated coding systems

    CN105122655A

  • Sliding window decoding method and system with retransmission mechanism

    CN112152642A

  • Sliding Window Method and Apparatus for Soft Input / Soft Output Processing

    US20080104488A1