A low-complexity method and system for optimizing a lossy protograph LDPC code

Abstract

The application relates to a low-complexity lossy protograph LDPC code optimization method and system, belonging to the technical field of LDPC code optimization. A base matrix of a protograph LDPC code is acquired first, then the degree of each column of the base matrix is calculated, the base matrix is adjusted in column position according to the degree of each column, a first matrix is obtained, then the first matrix is adjusted in parameters, a second matrix is obtained, finally, the second matrix is taken as a new base matrix, and an optimized protograph LDPC code is calculated, when the optimized protograph LDPC code is applied to a lossy source coding system, the rate distortion performance of the lossy source coding system can be effectively improved, meanwhile, the iteration times of coding and decoding are reduced, and the complexity of the lossy source coding system is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of LDPC code optimization, in particular to a structure optimization method and system of protograph LDPC code capable of reducing system complexity applied to lossy source coding. BACKGROUND

[0002] The definition of protograph LDPC (Low Density Parity Check) code was first proposed by J. Thorpe in his paper in 2003. Protograph LDPC code has low encoding and decoding complexity, excellent error correction rate and easy hardware implementation in theory, and therefore has attracted the attention and research of many scholars. However, it has been verified that when the traditional protograph LDPC code with excellent performance in channel coding is directly applied to lossy source coding, the lossy source coding system does not obtain the performance close to the theoretical rate-distortion bound, and the encoding and decoding iteration number is high.

[0003] Therefore, there is an urgent need for a structure optimization technology of protograph LDPC code. SUMMARY

[0004] The purpose of the present application is to provide a low-complexity lossy protograph LDPC code optimization method and system, which can effectively improve the rate-distortion performance of the lossy source coding system, reduce the encoding and decoding iteration number, and reduce the complexity of the lossy source coding system.

[0005] To achieve the above purpose, the present application provides the following scheme:

[0006] A low-complexity lossy protograph LDPC code optimization method, the optimization method comprising:

[0007] obtaining a base matrix of a protograph LDPC code;

[0008] calculating the degree of each column of the base matrix, and adjusting the column position of the base matrix according to the degree of each column to obtain a first matrix;

[0009] judging whether the difference between the number of rows and the number of columns of the first matrix is greater than or equal to a preset value;

[0010] if yes, adjusting the parameters of the first matrix to obtain a second matrix conforming to a first preset format;

[0011] if no, adjusting the parameters of the first matrix to obtain a second matrix conforming to a second preset format;

[0012] using the second matrix as a new base matrix to calculate an optimized protograph LDPC code;

[0013] the first preset format is: I is a unit matrix of m rows and m columns; 01 is a zero matrix of m-2 rows and n-m-1 columns; 1 is a matrix of 1 row and n-m-1 columns, and the elements of the matrix are 1; b mn is an element in the mth row and the nth column;

[0014] The second preset format is: I is a unit matrix of m rows and m columns; 01 is a zero matrix of m-2 rows and n-m-1 columns; 1 is a matrix of 1 row and n-m-1 columns, and the elements of the matrix are 1; b

[0015] In some embodiments, the calculating the degree of each column of the basis matrix specifically comprises: for each column of the basis matrix, summing all the elements of the column to obtain the degree of the column.

[0016] In some embodiments, the column position adjustment of the basis matrix according to the degree of each column to obtain a first matrix specifically comprises: arranging all the columns of the basis matrix in order from small to large degree to obtain a first matrix.

[0017] In some embodiments, the preset value is 2.

[0018] In some embodiments, the parameter adjustment of the first matrix to obtain a second matrix conforming to a first preset format specifically comprises:

[0019] For the first m columns of the first matrix, the element in the a-th column in the a-th row is adjusted to 1, and the rest of the elements are adjusted to 0, and at the same time, the adjustment amount of the i-th row of the a-th column is added to the i-th row of the last column of the first matrix; a=1, 2,..., m; i=1, 2,..., m;

[0020] For the middle columns of the first matrix, the elements in the b-th column in the m-1-th row and the m-th row are adjusted to 1, and the rest of the elements are adjusted to 0, and at the same time, the adjustment amount of the i-th row of the b-th column is added to the i-th row of the last column of the first matrix; b=m+1, m+2,..., n-1.

[0021] In some embodiments, the parameter adjustment of the first matrix to obtain a second matrix conforming to a second preset format specifically comprises:

[0022] For the first A columns of the first matrix, the element in the c-th column in the c-th row is adjusted to 1, and the rest of the elements are adjusted to 0, and at the same time, the adjustment amount of the i-th row of the c-th column is added to the i-th row of the last column of the first matrix; c=1, 2,..., A; i=1, 2,..., m;

[0023] For the middle column of the first matrix, the elements in the dth column at the m-1th row and the mth row are adjusted to 1, and the rest of the elements are adjusted to 0, while the adjustment amount of the ith row of the dth column is added to the ith row of the last column of the first matrix; d=A+1, A+2,..., n-1.

[0024] In some embodiments, after the optimized protograph LDPC code is obtained, the optimization method further comprises: encoding or decoding by using the optimized protograph LDPC code; and when the optimized protograph LDPC code is used for encoding or decoding, the last n-m columns of the second matrix are used as information bits.

[0025] A low-complexity lossy protograph LDPC code optimization system, the optimization system comprising:

[0026] A data acquisition module is configured to acquire a base matrix of a protograph LDPC code.

[0027] A position adjustment module is configured to calculate the degree of each column of the base matrix, and adjust the column positions of the base matrix according to the degree of each column to obtain a first matrix.

[0028] A parameter adjustment module is configured to determine whether the difference between the number of rows and the number of columns of the first matrix is greater than or equal to a preset value; if yes, the first matrix is adjusted in parameters to obtain a second matrix conforming to a first preset format; and if no, the first matrix is adjusted in parameters to obtain a second matrix conforming to a second preset format.

[0029] A structure optimization module is configured to take the second matrix as a new base matrix to calculate an optimized protograph LDPC code.

[0030] The first preset format is as follows: I is an m-row and m-column unit matrix; 01 is an m-2-row and n-m-1-column zero matrix; 1 is an 1-row and n-m-1-column matrix with elements of 1; and b mn is an element at the mth row and the nth column;

[0031] The second preset format is as follows: I is an A-row and A-column unit matrix; 02 is an m-A-row and A-column zero matrix; 01 is an m-2-row and n-A-1-column zero matrix; and 1 is an 1-row and n-A-1-column matrix with elements of 1.

[0032] According to the specific embodiments of the present application, the following technical effects are provided:

[0033] The application provides a low-complexity lossy protograph LDPC code optimization method and system, which comprises the following steps: obtaining a base matrix of a protograph LDPC code; calculating the degree of each column of the base matrix; adjusting the column position of the base matrix according to the degree of each column to obtain a first matrix; judging whether the difference between the number of rows and the number of columns of the first matrix is greater than or equal to a preset value; if yes, adjusting the parameters of the first matrix to obtain a second matrix conforming to a first preset format; if no, adjusting the parameters of the first matrix to obtain a second matrix conforming to a second preset format; and finally taking the second matrix as a new base matrix to calculate an optimized protograph LDPC code. When the optimized protograph LDPC code is applied to a lossy source coding system, the rate-distortion performance of the lossy source coding system can be effectively improved, the number of encoding and decoding iterations can be reduced, and the complexity of the lossy source coding system can be reduced. BRIEF DESCRIPTION OF DRAWINGS

[0034] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed in the embodiments. Obviously, the drawings described below are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative effort on the basis of these drawings.

[0035] Figure 1 The method flow chart of the optimization method provided by the embodiment 1 of the present application;

[0036] Figure 2 The rate-distortion performance comparison diagram of the original AR3A (Accumulate Repeat-3 Accumulate, cumulative repeat-3 accumulation) code and the optimized NAR3A-impr (New AR3A-impr, new cumulative repeat-3 accumulation) code provided by the embodiment 1 of the present application;

[0037] Figure 3 The rate-distortion performance comparison diagram of the original AR4JA (Accumulate Repeat-4 Jagged Accumulate, cumulative repeat-4 jagged accumulation) code and the optimized NAR4JA-impr (New AR4JA-impr, new cumulative repeat-4 jagged accumulation) code provided by the embodiment 1 of the present application;

[0038] Figure 4 The encoding process diagram provided by the embodiment 1 of the present application;

[0039] Figure 5 The system block diagram of the optimization system provided by the embodiment 2 of the present application. DETAILED DESCRIPTION

[0040] With reference to the accompanying drawings: clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the scope of the present application.

[0041] The purpose of the present application is to provide a low-complexity lossy protograph LDPC code optimization method and system, which can effectively improve the rate-distortion performance of the lossy source coding system, while reducing the number of iterations and the complexity of the lossy source coding system.

[0042] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application will be further described in detail below with reference to the drawings and specific embodiments.

[0043] Embodiment 1:

[0044] The present embodiment is to provide a low-complexity lossy protograph LDPC code optimization method, as shown in Figure 1 The optimization method comprises:

[0045] S1: obtaining a base matrix of a protograph LDPC code;

[0046] The present embodiment is to optimize the existing protograph LDPC code. In the optimization, the base matrix B of the protograph LDPC code is obtained, and the number of rows of the base matrix B is m and the number of columns is n.

[0047] S2: calculating the degree of each column of the base matrix, and adjusting the column position of the base matrix according to the degree of each column to obtain a first matrix;

[0048] In S2, the degree of each column of the base matrix can include: for each column of the base matrix, summing all elements of the column to obtain the degree of the column. The degree d vj of the jth column is calculated according to the formula: d vj = b 1j +b 2j +...+b mj , b ij is the element value of the ith row and jth column of the base matrix, i=1,2,...,m, j=1,2,...,n.

[0049] In S2, the size relationship between the degree d vj of the current column and the degree d vj+1 of the next column is judged column by column, and each time the degree d vj of the current column is greater than the degree d vj+1When the current column is replaced with the next column, the first matrix is generated after comparing all the columns. Specifically, the degree of each column is used to adjust the column position of the base matrix to obtain the first matrix, which can include arranging all the columns of the base matrix in ascending order of degree to obtain the first matrix. If the degrees of two columns are the same, the order of the two columns is arranged according to the original order.

[0050] S3: Determine whether the difference between the number of rows and the number of columns of the first matrix is greater than or equal to a preset value;

[0051] In this embodiment, the preset value can be 2, and S3 determines the size relationship between the number of rows m and the number of columns n of the first matrix. If n-m≥2 is satisfied, S4 is performed; if n-m<2 is satisfied, S5 is performed.

[0052] S4: If yes, the first matrix is adjusted to obtain a second matrix that meets a first preset format;

[0053] In this embodiment, the first preset format is: I is an m-row and m-column unit matrix; 01 is an m-2-row and n-m-1-column zero matrix; 1 is a 1-row and n-m-1-column matrix with elements of 1; b mn is the element in the mth row and the nth column.

[0054] Adjusting the first matrix to obtain a second matrix that meets the first preset format can include:

[0055] (1) For the first m columns of the first matrix, adjust the element in the a-th column in the a-th row to 1, and adjust the remaining elements to 0, while adding the adjustment amount of the i-th row of the a-th column to the i-th row of the last column of the first matrix; a=1, 2,..., m; i=1, 2,..., m; the adjustment amount is the difference between the adjusted element value and the initial element value of the i-th row of the a-th column.

[0056] (2) For the middle columns of the first matrix, adjust the elements in the b-th column in the m-1-th row and the m-th row to 1, and adjust the remaining elements to 0, while adding the adjustment amount of the i-th row of the b-th column to the i-th row of the last column of the first matrix; b=m+1, m+2,..., n-1.

[0057] This embodiment adjusts the number of columns with a degree of 1 in the first matrix. Specifically, the first m columns of the first matrix are adjusted to have a degree of 1, and the reduced amount is added to the corresponding row of the last column during adjustment, i.e., if b ij -q, then b in +q until the form of the matrix meets the first preset format to obtain the second matrix.

[0058] S5: If no, parameter adjustment is performed on the first matrix to obtain a second matrix conforming to a second preset format;

[0059] In this embodiment, the second preset format is: I is an A-row and A-column unit matrix, A is determined by the number of rows and columns of the original pattern and the layout characteristics of the matrix element values, and A < m; 02 is an m-A-row and A-column zero matrix; 01 is an m-2-row and n-A-1-column zero matrix; and 1 is an 1-row and n-A-1-column matrix with elements of 1.

[0060] The parameter adjustment performed on the first matrix to obtain a second matrix conforming to a second preset format can include:

[0061] (1) For the first A columns of the first matrix, the element in the c-th column at the c-th row is adjusted to 1, and the rest of the elements are adjusted to 0, and the adjustment amount of the i-th row of the c-th column is added to the i-th row of the last column of the first matrix; c = 1, 2,..., A; i = 1, 2,..., m.

[0062] (2) For the middle columns of the first matrix, the elements in the d-th column at the m-1-th row and the m-th row are adjusted to 1, and the rest of the elements are adjusted to 0, and the adjustment amount of the i-th row of the d-th column is added to the i-th row of the last column of the first matrix; d = A+1, A+2,..., n-1.

[0063] This embodiment adjusts the number of columns of degree 1 of the first matrix, specifically adjusts the first A columns of the first matrix to degree 1, and adds the reduced degree to the corresponding row of the last 1 column during adjustment, that is, if b ij -q, then b in +q, until the form of the matrix meets the second preset format, and a second matrix is obtained.

[0064] S6: Taking the second matrix as a new base matrix, an optimized original pattern LDPC code is calculated;

[0065] The process of calculating the optimized original pattern LDPC code according to the new base matrix is the existing process, which needs to generate a check matrix based on the new base matrix, and then generate the optimized original pattern LDPC code based on the check matrix.

[0066] After obtaining the optimized original pattern LDPC code, the optimization method of this embodiment further includes: encoding or decoding using the optimized original pattern LDPC code, and when encoding or decoding using the optimized original pattern LDPC code, the last n-m columns of the second matrix are used as information bits.

[0067] The optimization method corresponding to the AR3A code and the AR4JA code with a code rate of (n+1) / (n+2) (n>1) is the same as the optimization method of the AR3A code and the AR4JA code with a code rate of 1 / 2.

[0068] In this embodiment, the AR3A code with 1 / 2 code rate is taken as an example to further illustrate the optimization method.

[0069] Step 1: input the base matrix of the original protograph LDPC code The number of rows m=3 and the number of columns n=5.

[0070] Step 2: perform operations on the base matrix B in step 1 to generate the degrees of all columns (also referred to as variable nodes) of the base matrix B, and obtain d v1 =1, d v2 =5, d v3 =4, d v4 =2, d v5 =2.

[0071] Step 3: determine the size relationship between the degree d vj of the current column and the degree d vj+1 of the next column, and replace the elements of the current column with the elements of the next column whenever the degree d vj of the current column is greater than the degree d vj+1 of the next column, to obtain a first matrix, which is as follows:

[0072]

[0073] Step 4: determine the size relationship between the number of rows m and the number of columns n of the first matrix, and perform step 5 when n-m=5-3≥2.

[0074] Step 5: adjust the number of columns with degree 1 in the first matrix, and adjust the first m=3 columns of the first matrix to have degree 1, and add the reduced degrees to the corresponding rows of the last 1 column during the adjustment, i.e., b ij -1; b in +1, until the form of the matrix satisfies the following conditions:

[0075]

[0076] The obtained second matrix is as follows:

[0077]

[0078] Step 6: select the last n-m=5-3=2 columns of the second matrix as information bits, and output the base matrix of the optimized protograph LDPC code

[0079] In this embodiment, the AR3A code with 1 / 4 code rate is taken as an example to further illustrate the optimization method.

[0080] Step 1: AR3A code with channel coding rate of 1 / 4 is selected for optimization processing, when AR3A code with channel coding rate of 1 / 4 is used for source coding, its coding rate becomes 1 / 5, and the base matrix of its original modulo graph LDPC code is wherein the number of rows m=4, the number of columns n=5;

[0081] Step 2: the base matrix B is operated to calculate the degree of each column, and d v1 =1, d v2 =8, d v3 =2, d v4 =2, d v5 =1;

[0082] Step 3: the position of the column is adjusted according to the size of the degree of each column, when the degree d vj of the current column is greater than the degree d vj+1 of the next column, the elements of the current column and the elements of the next column are replaced, and the first matrix after adjustment is as follows:

[0083]

[0084] Step 4: the size relationship between the number of rows m and the number of columns n of the first matrix is judged, and m-n=1<2 is satisfied;

[0085] Step 5: the first matrix is parameter-adjusted to obtain a second matrix conforming to a second preset format as follows:

[0086]

[0087] Step 6: the last column is selected as information bits, and the base matrix of the optimized original modulo graph LDPC code is output as

[0088] The AR3A code and AR4JA code in the original modulo graph LDPC code are optimized by using the above optimization method, and the rate distortion performance before and after optimization is compared, and the results are as follows.

[0089] As Figure 2 shown, it is a rate distortion performance comparison diagram of the optimized NAR3A-impr code and the original AR3A code obtained by using the above optimization method, and it can be seen from the diagram that the rate distortion performance of the NAR3A-impr code is far superior to that of the AR3A code under the same code rate, and the performance close to the rate distortion theory boundary is obtained.

[0090] As Figure 3As shown in the figure, it is a comparison diagram of the rate-distortion performance of the optimized NAR4JA-impr code and the original AR4JA code obtained by using the above optimization method. As can be seen from the figure, the rate-distortion performance of the NAR4JA-impr code is better than that of the AR4JA code at the same code rate, and the performance close to the rate-distortion theory boundary is obtained.

[0091] As shown in Table 1, it is a comparison table of the iteration number between the AR3A code and the NAR3A-impr code at different source coding rates, including the iteration number of the optimized NAR3A-impr code and the system without optimization under the AR3A code at the corresponding code rate.

[0092] Table 1

[0093]

[0094] As can be seen from Table 1, the iteration number required by the system under the NAR3A-impr code is all below 70, and the iteration number required by the system under the AR3A code is as high as the maximum iteration number 300 set. Therefore, the optimized code has obvious advantages in the iteration number, thereby greatly reducing the system complexity.

[0095] As shown in Table 2, it is a comparison table of the iteration number between the AR4JA code and the NAR4JA-impr code at different source coding rates, including the iteration number of the optimized NAR4JA-impr code and the system without optimization under the AR4JA code at the corresponding code rate.

[0096] Table 2

[0097]

[0098] As can be seen from Table 2, the iteration number required by the system under the NAR4JA-impr code is all below 100, and the iteration number required by the system under the AR4JA code is as high as the maximum iteration number 300 set. Therefore, the optimized code also has obvious advantages in the iteration number.

[0099] In summary, the code structure of the AR3A code and the AR4JA code is optimized by using the optimization method of the embodiment, thereby improving the coding performance of the entire system, obtaining a better rate-distortion performance, and reducing the system complexity.

[0100] As Figure 4As shown, the encoding process applied to the optimized protograph LDPC code of the embodiment is introduced as follows: inputting the source sequence S at the input end of the encoder, applying the RBP (Reinforced Belief Propagation) algorithm to the check matrix H of the optimized protograph LDPC code, iteratively encoding the source sequence S to obtain the compressed sequence V, and transmitting the compressed sequence V to the decoder. After receiving the compressed sequence V, the decoder multiplies the compressed sequence V with the systematic form G of the generator matrix to recover the source sequence. L

[0101] Embodiment 2

[0102] The embodiment is used to provide a low-complexity lossy protograph LDPC code optimization system, as shown in the accompanying drawings, the optimization system comprises: Figure 5

[0103] A data acquisition module M1 is configured to acquire a base matrix of a protograph LDPC code.

[0104] A position adjustment module M2 is configured to calculate the degree of each column of the base matrix, and adjust the column positions of the base matrix according to the degree of each column to obtain a first matrix.

[0105] A parameter adjustment module M3 is configured to determine whether the difference between the number of rows and the number of columns of the first matrix is greater than or equal to a preset value. If yes, the parameter adjustment module M3 is configured to adjust the parameters of the first matrix to obtain a second matrix conforming to a first preset format. If no, the parameter adjustment module M3 is configured to adjust the parameters of the first matrix to obtain a second matrix conforming to a second preset format.

[0106] A structure optimization module M4 is configured to take the second matrix as a new base matrix, and calculate an optimized protograph LDPC code.

[0107] The first preset format is as follows: I is an m-row and m-column unit matrix; 01 is an m-2-row and n-m-1-column zero matrix; 1 is a 1-row and n-m-1-column matrix with elements of 1; b mn is the element in the mth row and the nth column;

[0108] The second preset format is as follows: I is an A-row and A-column unit matrix; 02 is an m-A-row and A-column zero matrix; 01 is an m-2-row and n-A-1-column zero matrix; 1 is a 1-row and n-A-1-column matrix with elements of 1.

[0109] ​​In the specification, each embodiment is mainly described in the difference from other embodiments, and the same or similar parts among the embodiments can be understood by referring to each other. For the system disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple, and the relevant parts can be understood by referring to the description of the method.

[0110] The principles and implementation manners of the present application are described by using specific examples in the specification, and the above description of the embodiments is only used to help understand the method of the present application and its core idea; meanwhile, for the general skilled in the art, the specific implementation manners and application ranges can be changed according to the idea of the present application. In summary, the content of the specification should not be understood as the limitation of the present application.

Claims

1. A low complexity method for optimizing a lossy protograph LDPC code, the method comprising: The optimization method comprises: acquiring a base matrix of a protograph LDPC code; calculating the degree of each column of the base matrix, and adjusting the column positions of the base matrix according to the degree of each column to obtain a first matrix; judging whether the difference between the number of rows and the number of columns of the first matrix is greater than or equal to a preset value; if yes, adjusting parameters of the first matrix to obtain a second matrix conforming to a first preset format; if no, adjusting parameters of the first matrix to obtain a second matrix conforming to a second preset format; taking the second matrix as a new base matrix to calculate an optimized protograph LDPC code; The first preset format is: ; I is an identity matrix of m rows and m columns; 01 is a zero matrix of m-2 rows and n-m-1 columns; 1 is a matrix of 1 row and n-m-1 columns, and elements of the matrix are 1; b mn is an element of the mth row and the nth column. The second preset format is: ; I is an A row and A column unit matrix; 02 is an m-A row and A column zero matrix; 01 is an m-2 row and n-A-1 column zero matrix; 1 is a 1 row and n-A-1 column matrix with elements of 1; the adjusting parameters of the first matrix to obtain the second matrix conforming to the first preset format specifically comprises: for the first m columns of the first matrix, adjusting the element in the a th column at the a th row to 1 and adjusting the rest elements to 0, and meanwhile adding the adjustment amount of the i th row of the a th column to the i th row of the last column of the first matrix; a = 1, 2,..., m; i = 1, 2,..., m; for the middle columns of the first matrix, adjusting the elements in the b th column at the m-1 th row and the m th row to 1 and adjusting the rest elements to 0, and meanwhile adding the adjustment amount of the i th row of the b th column to the i th row of the last column of the first matrix; b = m+1, m+2,..., n-1; the adjusting parameters of the first matrix to obtain the second matrix conforming to the second preset format specifically comprises: for the first A columns of the first matrix, adjusting the element in the c th column at the c th row to 1 and adjusting the rest elements to 0, and meanwhile adding the adjustment amount of the i th row of the c th column to the i th row of the last column of the first matrix; c = 1, 2,..., A; i = 1, 2,..., m; for the middle columns of the first matrix, adjusting the elements in the d th column at the m-1 th row and the m th row to 1 and adjusting the rest elements to 0, and meanwhile adding the adjustment amount of the i th row of the d th column to the i th row of the last column of the first matrix; d = A+1, A+2,..., n-1.

2. The optimization method of claim 1, wherein, the calculating the degree of each column of the base matrix specifically comprises: summing all the elements of each column of the base matrix to obtain the degree of the column.

3. The optimization method of claim 1, wherein, the adjusting the column positions of the base matrix according to the degree of each column to obtain the first matrix specifically comprises: arranging all the columns of the base matrix in the order of degree from small to large to obtain the first matrix.

4. The optimization method of claim 1, wherein, the preset value is 2.

5. The optimization method of claim 1, wherein, after obtaining the optimized protograph LDPC code, the optimization method further comprises: encoding or decoding by using the optimized protograph LDPC code; when encoding or decoding by using the optimized protograph LDPC code, taking the last n-m columns of the second matrix as information bits.

6. A low complexity lossy protograph LDPC code optimization system, comprising: the optimization system comprises: a data acquisition module configured to acquire a base matrix of a protograph LDPC code; a position adjustment module configured to calculate the degree of each column of the base matrix, and adjust the column positions of the base matrix according to the degree of each column to obtain a first matrix; The parameter adjustment module is configured to determine whether a difference between the number of rows and the number of columns of the first matrix is greater than or equal to a preset value; if yes, the first matrix is adjusted in parameters to obtain a second matrix conforming to a first preset format; and if no, the first matrix is adjusted in parameters to obtain a second matrix conforming to a second preset format. The structure optimization module is configured to take the second matrix as a new base matrix to calculate an optimized original mask LDPC code. The first preset format is: ; I is an identity matrix of m rows and m columns; 01 is a zero matrix of m-2 rows and n-m-1 columns; 1 is a matrix of 1 row and n-m-1 columns, and elements of the matrix are 1; b mn is an element of the mth row and the nth column. The second preset format is: ; I is an A row and A column unit matrix; 02 is an m-A row and A column zero matrix; 01 is an m-2 row and n-A-1 column zero matrix; 1 is a 1 row and n-A-1 column matrix with elements of 1; The parameter adjustment of the first matrix to obtain the second matrix conforming to the first preset format specifically includes: For the first m columns of the first matrix, an element in the a-th column and in the a-th row is adjusted to 1, and other elements are adjusted to 0, and meanwhile, an adjustment amount of the i-th row of the a-th column is added to the i-th row of the last column of the first matrix; a = 1, 2,..., m; i = 1, 2,..., m; For the middle columns of the first matrix, elements in the b-th column and in the m-1-th row and the m-th row are adjusted to 1, and other elements are adjusted to 0, and meanwhile, an adjustment amount of the i-th row of the b-th column is added to the i-th row of the last column of the first matrix; b = m+1, m+2,..., n-1. The parameter adjustment of the first matrix to obtain the second matrix conforming to the second preset format specifically includes: For the first A columns of the first matrix, an element in the c-th column and in the c-th row is adjusted to 1, and other elements are adjusted to 0, and meanwhile, an adjustment amount of the i-th row of the c-th column is added to the i-th row of the last column of the first matrix; c = 1, 2,..., A; i = 1, 2,..., m; For the middle columns of the first matrix, elements in the d-th column and in the m-1-th row and the m-th row are adjusted to 1, and other elements are adjusted to 0, and meanwhile, an adjustment amount of the i-th row of the d-th column is added to the i-th row of the last column of the first matrix; d = A+1, A+2,..., n-1.

Images