Fast decoding method and device based on maximum a posteriori probability soft decision

Abstract

The present invention provides a fast decoding method and apparatus based on maximum a posteriori probability soft decision, comprising constructing a probability calculation function based on a two-bit multi-base symbol sequence; obtaining the probability of all codewords in the block code; and initializing the number of iterations. item =0, construct all probability matrices for the current iteration number; set the current iteration number... item The probability matrix is ​​input into the probability calculation function to obtain the row probability vector and column probability vector corresponding to the probability matrix; the row probability vectors in the current iteration are added bit by bit to obtain the nth probability vector in the multi-base symbol sequence corresponding to the block code. k - item The probability of the position. k The length of the information bits; determine the number of iterations. item Is it less than k -2, if item < k -2, then update the probability matrix; if item = k If the value is -2, the iteration stops. This invention significantly reduces computational complexity, improves decoding efficiency, and reduces decoding overhead.

Description

Technical Field

[0001] This invention relates to the field of channel decoding technology, and in particular to a fast decoding method and apparatus based on maximum a posteriori probability soft decision. Background Technology

[0002] Block codes are an important and widely used type of coding in channel coding and decoding techniques. Maximum A Posteriori (MAP) soft-decision decoding is a decoding method with optimal decoding performance. However, MAP decoding algorithms have high computational complexity, and this problem is more pronounced in multi-ary systems, with a large amount of redundant computation, further increasing the computational complexity of the MAP algorithm.

[0003] Therefore, a technical solution is urgently needed to solve the above problems. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a fast decoding method and apparatus based on maximum a posteriori probability soft decision.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: On the one hand, this invention provides a fast decoding method based on maximum a posteriori probability soft decision, comprising the following steps: S1. Construct a probability calculation function based on a two-digit multi-base symbol sequence; S2. Obtain the probability of each codeword in the block code and initialize the number of iterations. item =0, construct all probability matrices for the current iteration number; S3, Set the current iteration number item The probability matrix is ​​input into the probability calculation function to obtain the row probability vector and column probability vector corresponding to the probability matrix; S4. Add the row probability vectors of the current iteration bit by bit to obtain the nth row in the multi-base symbol sequence corresponding to the block code. k - item The probability of the position. k The length of the information bits; S5. Determine the number of iterations. item Is it less than k -2, if item < k -2, then let item = item +1, update the probability matrix, and return S3; if item = k If the value is -2, then the iteration stops.

[0006] Furthermore, the probability calculation function is:

[0007]

[0008]

[0009]

[0010] in, For input parameters, one q × q Matrix; This is the row probability vector; It is a column probability vector; For input parameters Element; q Number in base; The first row in the probability vector w One element; The first column probability vector is the first column probability vector. e Each element.

[0011] Furthermore, the probability of each codeword in the block code is obtained according to the following formula:

[0012] in, For the first The probability of a code word; q Number in base; N The code length; For the first u The first code character j Individual character symbols; For the code sequence of the th j The bit code character is The input log-likelihood ratio.

[0013] Furthermore, construct all probability matrices for the current iteration number, including: Construct a vector from the probabilities of all codewords in the block code. ; vector Divided into Groups, each group contains There are elements such that the codeword corresponding to each group of elements is preceded by a certain number of elements. k - item -2 bits are the same; Arrange each group of elements according to the input format of the probability calculation function to obtain... probability matrices ,in, t is the index of the probability matrix.

[0014] Furthermore, in the multi-base symbol sequence corresponding to the block code, the first... k - item The probability of a given position is calculated using the following formula:

[0015] in, For the first in a multi-base symbol sequence k - item The probability of the position; t The index is the number of the probability matrix. The code character is At that time, the first item In the nth iteration t Row probability vectors corresponding to each probability matrix; q Integer numbers.

[0016] Furthermore, the probability matrix is ​​updated, including: make t The first digit of the codeword in the updated probability matrix corresponds to the first position of the permutation codeword. item The second bit of the codeword in the updated probability matrix is ​​used to rearrange the column probability vectors to form... The updated probability matrix.

[0017] On the other hand, the present invention provides a fast decoding apparatus based on maximum a posteriori probability soft decision, comprising: The first module is used to construct a probability calculation function based on a two-digit multi-base symbol sequence; The second module is used to obtain the probability of all codewords in the block code and initialize the number of iterations. item =0, construct all probability matrices for the current iteration number; The third module is used to set the current iteration number. item The probability matrix is ​​input into the probability calculation function to obtain the row probability vector and column probability vector corresponding to the probability matrix; The fourth module is used to sum the row probability vectors of the current iteration bit by bit to obtain the sign bit of the multi-base symbol sequence corresponding to the block code. k - item The probability of the position. k The length of the information bits; The fifth module is used to determine the number of iterations. item Is it less than k -2, if item < k -2, then let item = item +1, update the probability matrix, and return to the third module; if item = k If the value is -2, then the iteration stops.

[0018] Compared with the prior art, the beneficial technical effects of the present invention are as follows: The fast decoding method and apparatus based on maximum a posteriori probability soft decision provided by this invention calculates the probability of a constructed probability matrix by constructing a probability calculation function based on a two-bit multi-ary symbol sequence, thereby obtaining the nth bit in the multi-ary symbol sequence corresponding to the block code. k - item The probability of the digit is calculated, and then the probability matrix is ​​updated through iterative processing before continuing to calculate the probability of the digit in the multi-base symbol sequence. k - item The probability of +1 bit, until item = k -2, which means the probability of the first bit in the multi-base symbol sequence has been calculated, and the iteration stops.

[0019] The method described in this invention performs iterative calculations, reduces redundant calculations, thereby significantly reducing computational complexity, improving decoding efficiency, and reducing decoding costs. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0021] Figure 1 A flowchart of a fast decoding method based on maximum a posteriori probability soft decision is provided for one embodiment; Figure 2 This is a schematic diagram of the sumbox function calculation provided in one embodiment; Figure 3 This is a schematic diagram illustrating the information transfer between two iterations of the sumbox function on GF(4) in one embodiment; Figure 4 This is an iterative diagram of the fast decoding method GF(4) based on maximum a posteriori probability soft decision provided in one embodiment. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0023] ReferenceFigure 1 One embodiment provides a fast decoding method based on maximum a posteriori probability soft decision, including the following steps: S1. Construct a probability calculation function based on a two-digit multi-base symbol sequence; S2. Obtain the probability of each codeword in the block code and initialize the number of iterations. item =0, construct all probability matrices for the current iteration number; S3, Set the current iteration number item The probability matrix is ​​input into the probability calculation function to obtain the row probability vector and column probability vector corresponding to the probability matrix; S4. Add the row probability vectors of the current iteration bit by bit to obtain the nth row in the multi-base symbol sequence corresponding to the block code. k - item The probability of the position. k The length of the information bits; S5. Determine the number of iterations. item Is it less than k -2, if item < k -2, then let item = item +1, update the probability matrix, and return S3; if item = k If the value is -2, then the iteration stops.

[0024] By constructing a probability calculation function based on a two-bit multi-ary symbol sequence and performing probability calculations on the constructed probability matrix, the probability of the first bit in the multi-ary symbol sequence corresponding to the block code can be obtained. k - item The probability of the digit is calculated, and then the probability matrix is ​​updated through iterative processing before continuing to calculate the probability of the digit in the multi-base symbol sequence. k - item The probability of +1 bit, until item = k -2 indicates that the probability of obtaining the first bit in the multi-base symbol sequence has been calculated, and the iteration stops. By reducing redundant calculations through iterative calculations, the computational complexity is significantly reduced, the decoding efficiency is improved, and the decoding cost is reduced.

[0025] In a preferred embodiment, the probability calculation function is:

[0026]

[0027]

[0028]

[0029] in, For input parameters, one q × q Matrix; This is the row probability vector; It is a column probability vector; For input parameters The elements, each corresponding to a string of length 2. q The probability of a number with a given base and a sign; q Number in base; The first row in the probability vector w One element; The first column probability vector is the first column probability vector. e Each element.

[0030] The probability calculation function is denoted as the sumbox function. For input parameters, one q × q In a matrix, specifically, the columns correspond to the values ​​in the first sign bit, and the rows correspond to the values ​​in the second sign bit. The values ​​of the corresponding elements in both rows and columns range from 0 to... q -1, through this arrangement, makes the input parameters Able to systematically represent all possible lengths of 2 q The probability corresponding to the number base sign. Row probability vector. Sum of column probability vectors For two lengths q The vectors, respectively, correspond to the input parameters. The sum of elements in each row and the sum of elements in each column.

[0031] In one embodiment, the operation of the sumbox function on the finite field GF(4) is shown, such as Figure 2 As shown.

[0032] In a preferred embodiment, the probability of each codeword in the block code is obtained according to the following formula:

[0033] in, For the first The probability of a code word; q Number in base; N The code length; For the first u The first code character j Individual character symbols; For the code sequence of the th j The bit code character is The input log-likelihood ratio.

[0034] Construct all probability matrices for the current iteration number, including: Construct a vector from the probabilities of all codewords in the block code. ; vector Divided into Groups, each group contains There are elements such that the codeword corresponding to each group of elements is preceded by a certain number of elements. k - item -2 bits are the same; Arrange each group of elements according to the input format of the probability calculation function to obtain... probability matrices ,in, t is the index of the probability matrix. .

[0035] Update the probability matrix, including: make t The first digit of the codeword in the updated probability matrix corresponds to the first position of the permutation codeword. item The second bit of the codeword in the updated probability matrix is ​​used to rearrange the column probability vectors to form... The updated probability matrix.

[0036] The probability matrix is ​​updated using the method described above, and the column probability vectors are rearranged so that the next iteration calculates the nth probability vector in the multi-base symbol sequence. k - item The probability of +1 position.

[0037] In one embodiment, reference is made to Figure 3 This illustrates the information transfer process of the sumbox function on GF(4) between two iterations. In the figure, the column probability vector output by the sumbox function is shown. The rearranged matrix serves as the updated probability matrix. .

[0038] The multi-base symbol sequence corresponding to the block code is the first... k - item The probability of a given position is calculated using the following formula:

[0039] in, For the first in a multi-base symbol sequence k - item The probability of the position; t The index is the number of the probability matrix. The code character is At that time, the first item In the nth iteration t Row probability vectors corresponding to each probability matrix; qInteger numbers.

[0040] when item = k When -2, the probability of obtaining the first bit in the multi-base symbol sequence is: .

[0041] Reference Figure 4 One embodiment illustrates the iterative process on the fast decoding method GF(4) based on maximum a posteriori probability soft decision. The red part in the figure is the decoding output.

[0042] The method described in this invention is applicable to information of length [length missing]. k The block code, the first item The next iteration needs to be performed. The SUMBO function operation needs to be performed. k -1 iteration, totaling Each SumBox function operation includes... q 2 After several additions, it can be obtained that the method described in this invention is suitable for information of length [length missing]. k The number of additions required to perform the block code is

[0043] Existing multi-level MAP decoding algorithms require an accumulation computation of... , N Given the code length, it can be seen that the method described in this invention can reduce computational complexity. N times, by Descending to .

[0044] Therefore, this invention can significantly reduce computational complexity, improve decoding efficiency, and reduce decoding costs.

[0045] In one embodiment, a fast decoding apparatus based on maximum a posteriori probability soft decision is provided, comprising: The first module is used to construct a probability calculation function based on a two-digit multi-base symbol sequence; The second module is used to obtain the probability of all codewords in the block code and initialize the number of iterations. item =0, construct all probability matrices for the current iteration number; The third module is used to set the current iteration number. item The probability matrix is ​​input into the probability calculation function to obtain the row probability vector and column probability vector corresponding to the probability matrix; The fourth module is used to sum the row probability vectors of the current iteration bit by bit to obtain the sign bit of the multi-base symbol sequence corresponding to the block code. k- item The probability of the position. k The length of the information bits; The fifth module is used to determine the number of iterations. item Is it less than k -2, if item < k -2, then let item = item +1, update the probability matrix, and return to the third module; if item = k If the value is -2, then the iteration stops.

[0046] Matters not covered in this invention are common knowledge.

[0047] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0048] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these modifications and improvements all fall within the protection scope of this application.

[0049] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A fast decoding method based on maximum a posteriori probability soft decision, characterized in that, Includes the following steps: S1. Construct a probability calculation function based on a two-digit multi-base symbol sequence; S2. Obtain the probability of each codeword in the block code and initialize the number of iterations. item =0, construct all probability matrices for the current iteration number; S3, Set the current iteration number item The probability matrix is ​​input into the probability calculation function to obtain the row probability vector and column probability vector corresponding to the probability matrix; S4. Add the row probability vectors of the current iteration bit by bit to obtain the nth row in the multi-base symbol sequence corresponding to the block code. k - item The probability of the position. k The length of the information bits; S5. Determine the number of iterations. item Is it less than k -2, if item < k -2, then let item = item +1, update the probability matrix, and return S3; like item = k If the value is -2, then the iteration stops.

2. The fast decoding method based on maximum a posteriori probability soft decision as described in claim 1, characterized in that, The probability calculation function is: in, For input parameters, one q × q Matrix; This is the row probability vector; It is a column probability vector; For input parameters Element; q Number in base; The first row in the probability vector w One element; The first column probability vector is the first column probability vector. e Each element.

3. The fast decoding method based on maximum a posteriori probability soft decision as described in claim 1, characterized in that, The probability of each codeword in the block code is obtained according to the following formula: in, For the first The probability of a code word; q Number in base; N The code length; For the first u The first code character j Individual character symbols; For the code sequence of the th j The bit code character is The input log-likelihood ratio.

4. The fast decoding method based on maximum a posteriori probability soft decision as described in claim 3, characterized in that, Construct all probability matrices for the current iteration number, including: Construct a vector from the probabilities of all codewords in the block code. ; vector Divided into Groups, each group contains There are elements such that the codeword corresponding to each group of elements is preceded by a certain number of elements. k - item -2 bits are the same; Arrange each group of elements according to the input format of the probability calculation function to obtain... probability matrices ,in, t is the index of the probability matrix.

5. The fast decoding method based on maximum a posteriori probability soft decision as described in claim 1, characterized in that, The multi-base symbol sequence corresponding to the block code is the first... k - item The probability of a given position is calculated using the following formula: in, For the first in a multi-base symbol sequence k - item The probability of the position; t The index is the number of the probability matrix. The code character is At that time, the first item In the nth iteration t Row probability vectors corresponding to each probability matrix; q Integer numbers.

6. The fast decoding method based on maximum a posteriori probability soft decision as described in claim 4, characterized in that, Update the probability matrix, including: make t The first digit of the codeword in the updated probability matrix corresponds to the first position of the permutation codeword. item The second bit of the codeword in the updated probability matrix is ​​used to rearrange the column probability vectors to form... The updated probability matrix.

7. A fast decoding device based on maximum a posteriori probability soft decision, characterized in that, include: The first module is used to construct a probability calculation function based on a two-digit multi-base symbol sequence; The second module is used to obtain the probability of all codewords in the block code and initialize the number of iterations. item =0, construct all probability matrices for the current iteration number; The third module is used to set the current iteration number. item The probability matrix is ​​input into the probability calculation function to obtain the row probability vector and column probability vector corresponding to the probability matrix; The fourth module is used to sum the row probability vectors of the current iteration bit by bit to obtain the sign bit of the multi-base symbol sequence corresponding to the block code. k - item The probability of the position. k The length of the information bits; The fifth module is used to determine the number of iterations. item Is it less than k -2, if item < k -2, then let item = item +1, update the probability matrix, and return to the third module; if item = k If the value is -2, then the iteration stops.

Images