Search

CN-122001521-A - Method and device for encoding or decoding

CN122001521ACN 122001521 ACN122001521 ACN 122001521ACN-122001521-A

Abstract

The present application provides a method of encoding or decoding and a communication apparatus, in which a base matrix applied to LDPC is provided differently from a base matrix in the existing scheme. Compared with two columns with larger absolute values of the differences of the column weights in the base matrix of the prior proposal, the column weights of the two columns are more regular in the design of the base matrix provided by the application. Furthermore, the elements of the other row regions of the two columns outside the given row region are identical. When the two columns come from the column region of the base matrix meeting the corresponding characteristics and the given row region comes from the row region of the base matrix meeting the corresponding characteristics, the base matrix can generally reduce error floor under the application of different code lengths and code rates of a wireless system, so that the base matrix can be applied to a high-reliability communication scene.

Inventors

  • LIU KE
  • ZHANG HUAZI
  • TONG JIAJIE
  • WANG XIANBIN
  • QIN KANGJIAN
  • LI YUAN
  • WANG JUN

Assignees

  • 华为技术有限公司

Dates

Publication Date
20260508
Application Date
20241106

Claims (20)

  1. 1. A method of encoding, comprising: Acquiring a first base matrix, wherein the first base matrix corresponds to a second base matrix, the first base matrix comprises a first column and a second column, the first column and the second column respectively correspond to a third column and a fourth column of the second base matrix, the absolute value of the difference value of the column weights of the first column and the second column is smaller than the absolute value of the difference value of the column weights of the third column and the column weights of the fourth column, the second base matrix is a base matrix suitable for a first condition, the first condition is related to a code length and/or a code rate, the second base matrix and the first base matrix both comprise X rows and Y columns, the first column and the second column belong to a column set Q, the column set Q comprises at least two columns in the Y columns, the first column and the third column are identical in elements of other row areas outside a row area corresponding to a row set R, the second column and the fourth column are identical in elements of the other row areas outside the row area corresponding to the row set R, and the first column and the second column and the fourth column belong to an integer set Q; determining a first matrix according to the first base matrix and translation values corresponding to elements in the first base matrix; coding based on the first matrix to obtain a coded bit sequence, and Outputting the coded bit sequence.
  2. 2. A method of decoding, comprising: Acquiring a first base matrix, wherein the first base matrix corresponds to a second base matrix, the first base matrix comprises a first column and a second column, the first column and the second column respectively correspond to a third column and a fourth column of the second base matrix, the absolute value of the difference value of the column weights of the first column and the second column is smaller than the absolute value of the difference value of the column weights of the third column and the column weights of the fourth column, the second base matrix is a base matrix suitable for a first condition, the first condition is related to a code length and/or a code rate, the first base matrix and the second base matrix both comprise X rows and Y columns, the first column and the second column belong to a column set Q, the column set Q comprises at least two columns in the Y columns, other row areas of the first column and the third column outside a row area corresponding to a row set R are the same, the second column and the fourth column and the fourth column are at least one row of the same, and the other row areas of the first column and the second column and the fourth column are the same; determining a first matrix according to the first base matrix and translation values corresponding to elements in the first base matrix; decoding the received value sequence based on the first matrix to obtain a decoded bit sequence, and Outputting the coded bit sequence.
  3. 3. The method according to claim 1 or 2, wherein the obtaining a first base matrix comprises: And changing the second base matrix based on a first rule to obtain the first base matrix, wherein the first rule comprises exchanging elements of corresponding positions of the at least two columns contained in the column set Q in the row region corresponding to the row set R.
  4. 4. A method according to any one of claims 1 to 3, wherein the first base matrix comprises a sub-matrix a, a sub-matrix B, a sub-matrix C, a sub-matrix D and a sub-matrix E, wherein: The submatrix A is the 1 st to x1 st row and the 1 st to y1 st column of the first base matrix; the submatrix B is the 1 st to x1 st row and the y1+1 st to y2 nd column of the first base matrix; The submatrix C is the 1 st to the x1 st rows and the y2+1~Y th columns of the first base matrix; the submatrix D is the (x 1+ 1~X) th row, the (1-y 2) th column of the first base matrix; the submatrix E is the x1+1~X th row and y2+1~Y th column of the first base matrix; The row region corresponding to the row set R is located in the range of the x1+1~X row corresponding to the submatrix D, and X1 is less than or equal to 1 and less than or equal to X, Y1 is less than or equal to 1 and less than or equal to Y2, and X1, Y1 and Y2 are integers.
  5. 5. The method according to any one of claims 1 to 4, wherein said at least one row comprised by said set of rows R comprises a first row corresponding to a first threshold value being an upper threshold value of the number of columns belonging to said set of columns Q in said Y columns of said first basis matrix corresponding to said first row, said first threshold value being related to one or more of the following information: a current code rate; The row weight of the first row; The rows of the remaining columns of the first row except the columns that are punctured are heavy; the connection structure of the first row and the perforated columns comprises one of the following cases that none of the first row and the perforated columns are connected or that the first row is connected with at least one of the perforated columns.
  6. 6. The method according to any one of claims 1 to 5, wherein the row set R comprises at least two rows divided into at least two segments, the at least two segments being in one-to-one correspondence with at least two thresholds, wherein a threshold value corresponding to a second segment of the at least two segments indicates an upper threshold value of the number of columns belonging to the column set Q in the Y columns of the first base matrix corresponding to the segment, the second segment being any one of the at least two segments, and a first threshold value corresponding to the first row being a threshold value corresponding to the first segment.
  7. 7. The method of claim 6, wherein the index of the second segment is greater than the index of the first segment, the threshold value corresponding to the second segment is less than the threshold value corresponding to the first segment, and the first segment is any one of the at least two segments.
  8. 8. The method according to any one of claims 1 to 7, wherein the first column corresponds to a column index Y1 in the first base matrix, the third column corresponds to the column index Y1 in the second base matrix, 1+.y1+.y, Y1 is an integer; Wherein the third column has a column weight of d 1 in the row region of the second base matrix, a column weight of d ' 1 in other regions outside the row region, a column weight of d 2 in the row region of the first base matrix, a column weight of d' 2 ,d 1 in other regions outside the row region, and d ' 1 is greater than d' 2 .
  9. 9. The method of claim 8, wherein any one of the following is satisfied between d 1 ,d′ 1 ,d 2 and d' 2 : d 1 ＞d 2 and d' 1 ＞d′ 2 , or D 1 ＞d 2 and d' 1 ＞2d′ 2 , or D 1 ＝d 2 and d' 1 ＞d′ 2 , or D 1 ＝d 2 and d' 1 ＞2d′ 2 .
  10. 10. The method according to any one of claims 1 to 7, wherein the first column corresponds to a column index Y1 in the first base matrix, the third column corresponds to the column index Y1 in the second base matrix, 1+.y1+.x, Y1 is an integer; Wherein the third column has a column weight d 1 in the region other than the lower triangular parity row of the second base matrix, a column weight d ' 1 in the lower triangular parity row of the second base matrix and/or a row weight d ' 1 in the region other than the row region of the second base matrix, a column weight d 2 in the region other than the lower triangular parity row of the first base matrix, a column weight d ' 2 in the lower triangular parity row of the first base matrix and/or a row weight d ' 2 ,d 1 ＞d 2 and d ' 1 ＞d′ 2 in the region other than the row region of the first base matrix.
  11. 11. The method of claim 10, wherein d '1>2d'2.
  12. 12. The method according to any one of claims 4 to 11, wherein the column region corresponding to the column set Q is located within the range of the y1+1-y2 columns of the first base matrix.
  13. 13. The method according to any one of claims 4 to 12, wherein the set of columns Q comprises at least one column of a column region corresponding to the sub-matrix B of the first base matrix.
  14. 14. The method of claim 13, wherein at least one column in the column region corresponding to the sub-matrix B comprises a column with a column index of Y2, wherein the column with a column index of Y2 has an even column weight, or wherein the column with a column index of Y2 has a column with a minimum column weight in the column region corresponding to the sub-matrix B, and The column weight of the column with the column index of Y2 in the first base matrix is larger than the column weight of the column with the column index of Y2 in the second base matrix.
  15. 15. The method of claim 13 or 14, wherein the column set Q further comprises a column with a column index of Y3 and an odd column weight, and The column weight of the column with the column index of Y3 in the first base matrix is smaller than the column weight of the column with the column index of Y3 in the second base matrix.
  16. 16. The method according to any one of claims 4 to 11, wherein the column set Q comprises columns with even column weights in the column region corresponding to the sub-matrix B.
  17. 17. The method according to any one of claims 4 to 11, wherein the set of columns Q comprises at least one column in a column area corresponding to the sub-matrix a and at least one column in a column area corresponding to the sub-matrix B; And a column weight of at least one column in a column region corresponding to the submatrix a in the first base matrix is smaller than a column weight of at least one column in a column region corresponding to the submatrix a in the second base matrix; the column weight of at least one column in the column region corresponding to the sub-matrix B in the first base matrix is greater than the column weight of at least one column in the column region corresponding to the sub-matrix B in the second base matrix.
  18. 18. An encoding device, comprising: a processing module for: Acquiring a first base matrix, wherein the first base matrix corresponds to a second base matrix, the first base matrix comprises a first column and a second column, the first column and the second column respectively correspond to a third column and a fourth column of the second base matrix, the absolute value of the difference value of the column weights of the first column and the second column is smaller than the absolute value of the difference value of the column weights of the third column and the column weights of the fourth column, the second base matrix is a base matrix suitable for a first condition, the first condition is related to a code length and/or a code rate, the second base matrix and the first base matrix both comprise X rows and Y columns, the first column and the second column belong to a column set Q, the column set Q comprises at least two columns in the Y columns, the first column and the third column are identical in elements of other row areas outside a row area corresponding to a row set R, the second column and the fourth column are identical in elements of the other row areas outside the row area corresponding to the row set R, and the first column and the second column and the fourth column belong to an integer set Q; determining a first matrix according to the first base matrix and the translation values corresponding to the elements in the first base matrix, and Coding based on the first matrix to obtain a coding bit sequence; And the communication module is used for outputting the coded bit sequence.
  19. 19. A decoding apparatus, comprising: a processing module for: Acquiring a first base matrix, wherein the first base matrix corresponds to a second base matrix, the first base matrix comprises a first column and a second column, the first column and the second column respectively correspond to a third column and a fourth column of the second base matrix, the absolute value of the difference value of the column weights of the first column and the second column is smaller than the absolute value of the difference value of the column weights of the third column and the column weights of the fourth column, the second base matrix is a base matrix suitable for a first condition, the first condition is related to a code length and/or a code rate, the second base matrix and the first base matrix both comprise X rows and Y columns, the first column and the second column belong to a column set Q, the column set Q comprises at least two columns in the Y columns, the first column and the third column are identical in elements of other row areas outside a row area corresponding to a row set R, the second column and the fourth column are identical in elements of the other row areas outside the row area corresponding to the row set R, and the first column and the second column and the fourth column belong to an integer set Q; determining a first matrix according to the first base matrix and the translation values corresponding to the elements in the first base matrix, and Decoding the received value sequence based on the first matrix to obtain a decoded bit sequence; And the communication module is used for outputting the decoding bit sequence.
  20. 20. A communication device comprising a communication interface for obtaining information required to perform the method of any of claims 1,3-17 and for sending the information to a circuit for performing the method of any of claims 1,3-17 based on the information received, or The communication interface is configured to obtain information required to perform the method of any of claims 2-17 and to send the information to the circuit, which is configured to perform the method of any of claims 2-17 based on the information received.

Description

Method and device for encoding or decoding Technical Field The present application relates to the field of channel coding, and more particularly, to a method and apparatus for channel coding or decoding. Background Low density parity check code (LDPC) is a channel coding scheme very close to the shannon limit, and is determined as a data channel coding scheme of the fifth generation (the fifth generation, 5G) system. The LDPC used in the mainstream has a quasi-cyclic (QC) structure, and the QC-LDPC used in practice is represented by a Base Graph (BG). The base map may also be written in a matrix form, referred to as a base matrix. And lifting elements in the base matrix based on the lifting value, and expanding the base matrix into a complete check matrix for encoding or decoding. The 5G LDPC has poor performance when applied to low latency and high reliability (ultra-reliable and low-latency communication, URLLC), has an obvious error floor, and cannot support a high reliability scenario. In addition, since the code length, code rate, etc. of the wireless system are very flexible, there are also very many combinations of different code lengths and code rates, and if each combination is independently optimized, the difficulty of optimization is great. Therefore, in a high-reliability scene, how to generally improve the performance of the LDPC at different code lengths and code rates is a problem to be solved. Disclosure of Invention The application provides a coding or decoding method and a device, which are applied to LDPC and can cause the error floor under different code lengths and code rates to be generally reduced. In a first aspect, an encoded method is provided that may be performed by a communications device or a module (e.g., a processor, chip, circuit, etc., or logic module, hardware, and/or software that may implement all or part of the functionality of the communications device) applied to the communications device. The communication device is also referred to as an encoding device. Acquiring a first base matrix, wherein the first base matrix corresponds to a second base matrix, the first base matrix comprises a first column and a second column, the first column and the second column respectively correspond to a third column and a fourth column of the second base matrix, the absolute value of the difference value of the column weight of the first column and the column weight of the second column is smaller than the absolute value of the difference value of the column weight of the third column and the column weight of the fourth column, the second base matrix is a base matrix applicable to a first condition, the first condition is related to code length and/or code rate, the second base matrix and the first base matrix both comprise X rows and Y columns, the first column and the second column belong to a column set Q, the column set Q comprises at least two columns in the Y columns, the first column and the third column are identical in elements of other row areas outside a row area corresponding to a row set R, the second column and the fourth column and the row set Q comprise at least one integer of elements of the other row areas outside the row area corresponding to the row R; determining a first matrix according to the first base matrix and the translation value corresponding to the element in the first base matrix, coding based on the first matrix to obtain a coding bit sequence, and outputting the coding bit sequence. In this technical scheme, the base matrix applied to LDPC is different from the base matrix in the existing scheme (e.g., BG1 or BG2 in NR). Compared with two columns with larger absolute values of the differences of the column weights in the base matrix of the prior proposal, the column weights of the two columns are more regular in the design of the base matrix provided by the application. Furthermore, the elements of the other row regions of the two columns outside the given row region are identical. When the two columns come from the column region of the base matrix meeting the corresponding characteristics and the given row region comes from the row region of the base matrix meeting the corresponding characteristics, the base matrix can generally reduce error floor under the application of different code lengths and code rates of a wireless system, so that the base matrix can be applied to a high-reliability communication scene. For the features that should be satisfied by the column region or the row region, reference may be made to the following various implementations or implementations in the specific embodiments, which are not described herein. In the embodiment of the present application, the numbers of the "first column", "second column", etc. are columns meeting certain requirements in the first base matrix, and the numbers of the "first column", "second column" are only for clarity of description, and do not represent the first column or the second column of t