(完整版)MatlabLDPC码性能研究说明书毕业设计论文

优秀论文 审核通过 未经允许 切勿外传 本科毕业设计（论文）

题 目 LDPC码性能研究

摘 要

信道编码是数字通信系统的重要组成部分。LDPC信道编码技术是编

码界的重要成果之一。12码率的二元LDPC码在AWGN信道下的性能距信息理论中的Shannon极限仅差0.0045dB，它是目前距Shannon极限最近的纠错码。Gallagar在1962年提出LDPC码，1996年经过Mackey、Spielman和Wiberg等人的再发现后，LDPC码以其性能优越、全并行迭代译码结构、便于硬件实现等优点，在无线通信、存储工业等领域得到了广泛应用。

校验矩阵的构造是编码的前提，本文采用了随机构造法构造，并对矩

阵的多种变换方法进行分析，比较了优缺点。译码算法是LDPC码的关键，译码复杂度的大小直接影响系统的实现。主要分硬判决译码、软判决和复合译码。设计中采用的译码方式是软判决译码。

在性能分析方面，利用matlab仿真码长、列重和迭代次数对LDPC码

性能的影响。仿真结果表明，在一定范围内，LDPC码长码的误码性能优于短码；码长较小时，列重的增加会使性能变差，而对于长码，列重在一定范围内的增大会改善LDPC的误码性能；增加迭代次数会使误码率降低，但当迭代次数大到一定值时，误码率将不会再随着迭代次数的增加而降低。

关键词 LDPC码 信道编码 矩阵变换 性能分析

Abstract

Channel coding is an important component for digital communication

systems. LDPC channel coding technology is one important achievement of the encoding results. The performance of AWGN channel only Limit that makes it be the latest error-correcting codes from the Shannon Limit. Gallager proposed LDPC codes in 1962, after Mackey and others re-discovered it in 1996, with best performance, completely decoding algorithm in parallel scheme and easily realized for , LDPC codes widely used in many practical systems such as wireless communication system and storage system.

The construction of the Check matrix is a precondition for coding; the randomized method was used in this paper. Several ways to transform the matrix was discussed and being compared about the merits of each method. The decoding algorithm is the key to LDPC code and the complexity of decoding direct impact realization of the system. There are three kinds of decoding algorithms: method, soft-decision methods and method was used in this design.

Based on studying the basic theory of LDPC codes, the impact of codes length, column weight and iteration times on BER performance of LDPC codes are demonstrated by computer simulation and theoretical analysis. The simulation experiment indicated that: LDPC codes long-code error performance is superior to short-code error performance, but when the code length reaches a certain value, and then increase the code length, BER of LDPC codes lower rate will be small. When the code length is small, increasing the column weight, LDPC codes performance will deteriorate; when code length is large enough, increasing the column weight, LDPC codes performance will be improved, but when the column weight reaches