Reconstruction of sparse check matrix for LDPC at high bit error rate

In order to reconstruct the sparse check matrix of LDPC, a new algorithm which could directly reconstruct the LDPC was proposed.Firstly, according to the principle of the traditional reconstruction algorithm, the defects of the traditional algorithm and the reasons for the defects were analyzed in d...

Full description

Saved in:
Bibliographic Details
Main Authors: Zhaojun WU, Limin ZHANG, Zhaogen ZHONG, Renxin LIU
Format: Article
Language:zho
Published: Editorial Department of Journal on Communications 2021-03-01
Series:Tongxin xuebao
Subjects:
Online Access:http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2021009/
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In order to reconstruct the sparse check matrix of LDPC, a new algorithm which could directly reconstruct the LDPC was proposed.Firstly, according to the principle of the traditional reconstruction algorithm, the defects of the traditional algorithm and the reasons for the defects were analyzed in detail.Secondly, based on the characteristics of sparse matrix, some bit sequences in code words were randomly extracted for Gaussian elimination.At the same time, in order to reliably realize that the extracted bits sequence could contain parity check nodes, the multiple random variables were determined based on the probability of containing check nodes in one extraction.Finally, the statistical characteristics of LDPC under the suspected check vector was analyzed.Based on the minimum error decision rule, the sparse check vector was determined.The simulation results show that the rate of reconstruction of most LDPC in IEEE 802.11 protocol can reach more than 95% at BER of 0.001, and the noise robustness of the proposed method is better than that of the traditional algorithm.At the same time, the new algorithm not only does not need sparseness of parity check matrix, but also has the good performance for both diagonal and non-diagonal check matrix.
ISSN:1000-436X