Search

CN-122026925-A - Communication method and communication device based on LDPC code

CN122026925ACN 122026925 ACN122026925 ACN 122026925ACN-122026925-A

Abstract

The utility model provides a communication method and communication device based on LDPC, this method can be based on the LDPC matrix and encode or decode, the LDPC matrix is based on LDPC code translation value matrix, LDPC code translation value matrix includes the partial or whole area of first translation value matrix, first translation value matrix includes first region and second area, wherein, first region is the matrix that 1 element in the partial or whole area of LDPC base matrix was replaced corresponding translation value, 1 element's corresponding translation value in the base matrix is based on the value Zc and determines, including k 1X regions in the first region, the second region includes k 2Y regions, the translation value of each regional and the translation value of a regional in k 1X region in k 2Y regions satisfy a rule, and k 2Y regions correspond different rules. The method can improve the performance of the LDPC code in a high throughput scene.

Inventors

  • Lv Zequn
  • ZHANG HUAZI
  • LI YUAN
  • LIU KE
  • WANG JUN
  • WANG XIANBIN
  • TONG JIAJIE

Assignees

  • 华为技术有限公司

Dates

Publication Date
20260512
Application Date
20241111

Claims (20)

  1. 1. A communication method based on low density parity check, LDPC, codes, the method comprising: Acquiring an information bit sequence; encoding the information bit sequence according to the LDPC matrix to obtain a codeword sequence, Wherein the LDPC matrix is determined based on an LDPC code shift value matrix and a boost value Zc, the LDPC code shift value matrix comprising a partial or full region of a first shift value matrix comprising a first region and a second region, The first area is a matrix obtained by replacing 1 element in part or all of the areas of the LDPC base matrix with a translation value corresponding to the 1 element, the translation value corresponding to the 1 element in the LDPC base matrix is determined based on the Zc, the first area comprises k 1X areas, the second area comprises k 2Y areas, the k1 is smaller than or equal to the k2, The translation value of the ith row of any one of the k 2Y regions Y q and the translation value of the ith row of any one of the k 1X regions X p satisfy a first rule, the region X p is a region corresponding to the region Y q , the ith row is any one of the regions Y q , and the ith row corresponds to the ith row.
  2. 2. A communication method based on low density parity check LDPC code is characterized in that, Acquiring a symbol sequence; decoding the symbol sequence according to the LDPC matrix to obtain an information bit sequence, Wherein the LDPC matrix is determined based on an LDPC code shift value matrix and a boost value Zc, the LDPC code shift value matrix comprising a partial or full region of a first shift value matrix comprising a first region and a second region, The first area is a matrix obtained by replacing 1 element in part or all of the areas of the LDPC base matrix with a translation value corresponding to the 1 element, the translation value corresponding to the 1 element in the LDPC base matrix is determined based on the Zc, the first area comprises k 1X areas, the second area comprises k 2Y areas, the k1 is smaller than or equal to the k2, The translation value of the ith row of any one of the k 2Y regions Y q and the translation value of the ith row of any one of the k 1X regions X p satisfy a first rule, the region X p is a region corresponding to the region Y q , the ith row is any one of the regions Y q , and the ith row corresponds to the ith row.
  3. 3. The method according to claim 1 or2, wherein a translation value q t,i,j of an i 'th row and j' th column of region Y q and a translation value p t,i,j of an i 'th row and j' th column of region X p satisfy the first rule, wherein the j 'th column and the j' th column are columns of the i 'th row and the i' th row.
  4. 4. A method according to claim 3, wherein the first rule relates to a value corresponding to the i-th row of the region X p .
  5. 5. The method of claim 4, wherein the first rule satisfies the following formula: q t,i,j ＝p t,i,j +β t,i , Or alternatively, the first and second heat exchangers may be, Q t,i,j ＝p t,i,j +β t,i h (Zc), said h (Zc) being a function related to said Zc, Or alternatively, the first and second heat exchangers may be, q t,i,j ＝β t,i *p t,i,j , Wherein β t,i is the value corresponding to the i-th row.
  6. 6. The method of claim 4, wherein the first rule satisfies the following formula q t,i,j ＝α t,i *p t,i,j +β t,i , wherein α t,i and β t,i are values corresponding to the ith row.
  7. 7. The method of claim 4, wherein the first rule satisfies the formula q t,i,j ＝f t,i (p t,i,j . The f t,i is a nonlinear function corresponding to at least one constant for nonlinear transformation, the at least one constant for nonlinear transformation being a value corresponding to the i-th line.
  8. 8. The method of any one of claims 4 to 7, wherein the ith row of the region X p is located on the kth row in the first region, and the value corresponding to the ith row of the region X p is a value in a sequence corresponding to the kth row in a predefined plurality of sequences, the plurality of sequences including a sequence corresponding to each row in the first region.
  9. 9. The method of any one of claims 1 to 8, wherein the k1 is equal to the k2, and the translation value of region Y t in the k 2Y regions and the translation value of region X t in the k 1X regions satisfy a rule t, 1+.t+.k1.
  10. 10. The method according to any one of claims 1 to 9, wherein the first area corresponds to a first matrix obtained by replacing the value of a position in the first area where a translation value exists with 1 element and replacing the value of the remaining position with 0 element, the first area including X 'rows and Y' columns, wherein, The matrix formed by the x1'+1~X' row and the y2'+1~Y' column of the first matrix is an identity matrix, The matrix formed by the 1 st to x1 st rows and the y2' +1~Y th columns of the first matrix is an all-0 matrix, The matrix formed by the 1 st to x1' rows and the y1' +1 to y2' columns of the first matrix is a square matrix, Wherein 1< X1' < X ',1< Y1' < Y2' < Y ', X1, X ', Y1', Y2', Y ' are integers.
  11. 11. The method of claim 10, wherein the step of determining the position of the first electrode is performed, The first region is composed of a third region and a fourth region, the fourth region is a region composed of at least one column in the first region, and each region in the k 1X regions is a region composed of at least one row of the third region.
  12. 12. The method of claim 11, wherein the rows in which the k 1X regions are located are each consecutive rows.
  13. 13. The method according to claim 11 or 12, wherein, The fourth region is a region constituted by Y2'+1 to Y' columns in the first region, Or alternatively, the first and second heat exchangers may be, The fourth region is a region formed by the y1'+1~Y' th column in the first region.
  14. 14. The method of claim 13, wherein the LDPC base matrix is base figure 1, the first matrix is the base figure 1, One of the k 1X regions is a region constituted by the first 4 rows in the third region, The other region of the k 1X regions is a region formed by the 5 th line to the 22 nd line of the third region, or the other region of the k 1X regions is a region formed by the 5 th line to the 24 th line of the third region, The other region of the k 1X regions is a region constituted by the 25 th to 46 th rows of the third region.
  15. 15. The method according to any one of claims 11 to 14, wherein, The first translation value matrix is composed of the third area, the second area and the fourth area, wherein the row numbers of the second area and the third area are the same, the last column of the third area and the first column of the second area are adjacent columns, and the last column of the second area and the first column of the fourth area are adjacent columns.
  16. 16. The method of claim 15, wherein the number of columns in the second region and the third region are the same.
  17. 17. The method of claim 16, wherein the region Y q is the same size as the region X p .
  18. 18. The method of claim 17, wherein the location of the region Y q in the second region and the location of the region X p in the third region are the same.
  19. 19. The method of claim 18, wherein the region Y q is the same as the location in the region X p where there is a translation value, and the translation value at the same location as in the region X p satisfies the first rule.
  20. 20. The method of any one of claims 1 to 19, wherein Zc is greater than or equal to 64.

Description

Communication method and communication device based on LDPC code Technical Field The present application relates to the field of encoding, and more particularly, to a communication method and a communication apparatus based on an LDPC code. Background In the field of channel coding, low Density Parity Check (LDPC) codes are one of the most well-established and widely-used channel coding schemes. Quasi-cyclic low-density parity check (QC-LDPC) codes are a type of structured LDPC codes, and due to the unique structure of the check matrix, the encoding can be realized by using a simple feedback shift register, so that the encoding complexity of the LDPC codes is reduced. In the field of modern communication, an efficient and reliable channel coding scheme has a key meaning for guaranteeing the performance and efficiency of information transmission. The existing LDPC code has low convergence speed under the high throughput scene and poor performance under the low iteration number, so how to improve the performance of the LDPC code under the low iteration number becomes a problem to be solved urgently. Disclosure of Invention The embodiment of the application provides a communication method and a communication device based on an LDPC code, which can improve the performance of the LDPC code in a high throughput scene. In a first aspect, a communication method based on an LDPC code is provided, where the method may be performed by a sender device, where the "sender device" in the present application may refer to the sender device itself (e.g. a network device, a terminal device), a component in the sender device (e.g. a processor, a chip, or a chip system, etc.), or may be a logic module or software capable of implementing all or part of the functions of the sender device, where the method is not specifically described. The method comprises the steps of obtaining an information bit sequence, encoding the information bit sequence according to an LDPC matrix to obtain a codeword sequence, wherein the LDPC matrix is determined based on an LDPC code translation value matrix and a lifting value Zc, the LDPC code translation value matrix comprises a part or all of areas of a first translation value matrix, the first translation value matrix comprises a first area and a second area, the first area is a matrix obtained by replacing 1 element in the part or all of areas of the LDPC matrix with a translation value corresponding to 1 element, the translation value corresponding to 1 element in the LDPC matrix is determined based on Zc, the first area comprises k 1X areas, the second area comprises k 2Y areas, k1 is smaller than or equal to k2, the translation value of an ith row of any area Y q in the k 2Y areas and the translation value of an ith row of any area X p in the k 1X areas meet a first rule, the area X p is an area corresponding to the area Y q, the translation value corresponding to the 1 element in the LDPC matrix is determined based on Zc, the first area comprises k 1X areas, the ith row corresponds to the ith row. Wherein the translation value of the ith row of any one of the k 2Y regions Y q and the translation value of the ith row of any one of the k 1X regions X p satisfy a first rule, it can be understood that the translation value of each of the k 2Y regions and the translation value of one of the k 1X regions satisfy a rule. Alternatively, the rules corresponding to different regions in the k 2Y regions may be the same or different. Optionally, the rule corresponding to each of the k 2Y regions is different. For example, k1=k2, and the translation value of region Y t in k 2Y regions and the translation value of region X t in k 1X regions satisfy the rule t, 1+.t+.k1. Illustratively, at least two of the k1< k2, k 2Y regions correspond to one of the k 1X regions. For example, k 2Y regions are regions Y 1、Y2、Y3、Y4 and Y 5, k 1X regions are regions X 1、X2 and X 3, wherein the translation values of region Y 1 and region X 1 satisfy rule #1, the translation values of region Y 2 and region X 1 satisfy rule #2, the translation values of region Y 3 and region X 2 satisfy rule #3, the translation values of region Y 4 and region X 2 satisfy rule #4, and the translation values of region Y 5 and region X 3 satisfy rule #5. In the above technical solution, the first area may be regarded as a translation value matrix obtained based on the base matrix, and the second area may be regarded as a translation value matrix obtained based on the first area. In order to improve the performance of the LDPC code under the low iteration number, the number of columns of the translation value matrix needs to be increased, and then the effect of increasing the number of columns of the translation value matrix can be achieved based on the second region in the application. In addition, the translation value of the second area is obtained based on the same method, so that the LDPC code performance is poor, the first area is divided int