一种改进的LDPC码低复杂度最小和算法  

Improved Min-sum Algorithm with Low Complexity for Decoding LDPC Codes

作　　者：吴军[1] 廖鑫[1] 张小红[1] 

机构地区：[1]江西理工大学信息工程学院,江西赣州341000

出　　处：《电视技术》2015年第1期88-91,95,共5页Video Engineering

摘　　要：研究了低密度奇偶校验(Low-Density Parity-Check,LDPC)码的单最小值最小和(Single-Minimum Min-Sum,SMMS)算法,为了提高译码性能,在此基础上提出一种信道自适应可配置LDPC码最小和译码(Adaptive Configurable Min-Sum,ACMS)算法。ACMS算法在BP译码时的横向消息迭代更新过程中,LLR次小值用一个基于迭代次数的估算参数与最小值相加来取代,同时根据每次判决时的错误比特个数对不同信噪比下的估算参数进行动态修正。仿真结果表明,ACMS算法整体上提高了译码性能而仅增加少量复杂度。In order to reduce the performance loss of SMMS （ Single-Minimum, Min-Sum） algorithm for decoding LDPC （ Low- Density Parity-Check） codes, the ACMS （adaptive configurable Min-Sum） algorithm is proposed in this paper. Only the absolute minimum is used in this algorithm to update the check-node message and the second minimum is determined by a configurable estimation weight factor based on the iteration and a correction factor based on the number of error bits in each hard decision. Compared to the original SMMS and VWMS algorithm, the simulation result shows the proposed ACMS algorithm reduce the performance loss in general by introducing only a little additional complexity.

关 键 词：低密度奇偶校验码 最小和算法 单最小值 估算参数 信道自适应 

分 类 号：TN911.22[电子电信—通信与信息系统]