Performance analysis of finite-length LDPC codes

Author

Xue, Yingjian ; Xiang, Haige

Author_Institution

Dept. of Electron., Peking Univ., Beijing, China

fYear

2003

fDate

15-18 June 2003

Firstpage

85

Lastpage

89

Abstract

In this paper, we are concerned with the finite-length analysis of low-density parity-check (LDPC) codes when used over additive white Gaussian noise (AWGN) channels. We present the iterative decoding algorithm of LDPC codes, and we also discuss the encoding methods and the bipartite graph representation of LDPC codes ensemble. Finite-length LDPC code performances are investigated, and the influence of cycle is pointed out. Simulation results show that irregular LDPC codes with variable and check node degree distribution optimized for infinite-length codes are not optimal for finite-length LDPC codes, and the error patterns of both regular and irregular codes are investigated as an effort to improve the performance and design methodology of finite-length LDPC codes.

Keywords

AWGN channels; error detection codes; graph theory; iterative decoding; parity check codes; AWGN channels; additive white Gaussian noise channels; bipartite graph representation; cycle influence; error patterns; finite-length LDPC codes; irregular LDPC codes; iterative decoding algorithm; low-density parity-check codes; performance analysis; AWGN; Additive white noise; Belief propagation; Design methodology; Iterative algorithms; Iterative decoding; Maximum likelihood decoding; Parity check codes; Performance analysis; Turbo codes;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing Advances in Wireless Communications, 2003. SPAWC 2003. 4th IEEE Workshop on

Print_ISBN

0-7803-7858-X

Type

conf

DOI

10.1109/SPAWC.2003.1318927

Filename

1318927