基于二元译码信息的迭代大数逻辑LDPC译码算法及其量化优化

发布时间：2018-03-21 04:39

本文选题：译码算法　切入点：LDPC码　出处：《电子与信息学报》2017年04期 　论文类型：期刊论文

【摘要】：该文提出一种低复杂度的迭代大数逻辑LDPC译码算法,在迭代过程中所有的译码信息都以二元形式进行传递、处理和迭代更新。所提算法不需要计算外信息,而是利用Tanner图上伴随式的对错状态来评判节点可靠度。与现有的几种迭代大数逻辑译码算法相比,该文算法也不需要信息修正处理,避免了相应的实数乘法操作,具有很低的译码复杂度。此外,该文引入一种特殊的量化处理函数,并给出了基于离散密度进化的参数优化过程。实验仿真表明,该文所提算法与原算法相比,在AWGN信道下可获得约0.3~0.4 d B的性能提升。同时,由于节点间交换传递的译码信息都是基于1个比特位的二元信息,也非常便于硬件的设计与实现。
[Abstract]:This paper proposes a low complexity iterative majority logic decoding algorithm of LDPC in the process of iterative decoding information all in the form of two yuan of transfer, update processing and iteration. The proposed algorithm does not need to compute the information, but the use of Tanner map with the right to judge the state of node reliability. Compared with several iterative majority logic decoding algorithms available, this algorithm does not need to modify information processing, to avoid the corresponding number of multiplication operations, with very low decoding complexity. In addition, this paper introduces a special quantization processing function, and gives the process parameter optimization of discrete density evolution based on experiments. Simulation results show that the proposed algorithm can acquire about 0.3~0.4 D B in AWGN channel performance. At the same time, because the decoding information transfer between nodes is $two based on 1 bits of information, It is also very convenient for the design and implementation of hardware.

【作者单位】：华南理工大学电子与信息学院;广西大学计算机与电子信息学院;广西多媒体通信与网络技术重点实验室培育基地/广西高校多媒体通信与信息处理重点实验室;
【分类号】：TN911.22

【相似文献】