Entropy (Aug 2023)

Design of Low-Density Parity-Check Code Pair for Joint Source-Channel Coding Systems Based on Graph Theory

  • Yijie Lv,
  • Jiguang He,
  • Weikai Xu,
  • Lin Wang

DOI
https://doi.org/10.3390/e25081189
Journal volume & issue
Vol. 25, no. 8
p. 1189

Abstract

Read online

In this article, a graph-theoretic method (taking advantage of constraints among sets associated with the corresponding parity-check matrices) is applied for the construction of a double low-density parity-check (D-LDPC) code (also known as LDPC code pair) in a joint source-channel coding (JSCC) system. Specifically, we pre-set the girth of the parity-check matrix for the LDPC code pair when jointly designing the two LDPC codes, which are constructed by following the set constraints. The constructed parity-check matrices for channel codes comprise an identity submatrix and an additional submatrix, whose column weights can be pre-set to be any positive integer numbers. Simulation results illustrate that the constructed D-LDPC codes exhibit significant performance improvement and enhanced flexible frame length (i.e., adaptability under various channel conditions) compared with the benchmark code pair.

Keywords