Improving the Decoding Performance of High-Rate GLDPC Codes in Low Error-Rate Applications

Abstract

This paper presents a new evaluation and performance comparison of a reliability-based iterative decoder for Generalized LDPC codes with BCH subcodes, using Soft-Input Soft-Output (SISO) Chase algorithm, compared with the same algorithm employing Hamming subcodes. While the single-error correction code with length n is accorded a preference over the double error correction one with length 2n (in terms of error performance over AWGN channels), this article shows that the GLDPC code with BCH subcodes surpasses the Hamming-based one under nearly the same overall code rate. It also converges in less than half the number of iterations compared to different corresponding Hamming-based soft-decision decoding (SDD) systems. As a case study, Gallager-based global LDPC code is applied and the decoder is considered over AWGN channel. The simulation results are showing the performance against the code block length, number of iterations and Chase algorithm parameters.