LDPC码的分层自适应最小和译码算法  

Hierarchical adaptive minimum sum decoding algorithm for LDPC code

作　　者：郑仁乐 李东阳 刘文学 万金涛 刘学勇[1,2] 李金海 ZHENG Renle;LI Dongyang;LIU Wenxue;WAN Jintao;LIU Xueyong;LI Jinhai(Communication and Information Engineering Research and Development Center,Institute of Microelectronics of the Chinese Academy of Sciences,Beijing 100029,China;School of Integrated Circuits,University of Chinese Academy of Sciences,Beijing 100049,China)

机构地区：[1]中国科学院微电子研究所通信与信息工程研发中心,北京100029 [2]中国科学院大学集成电路学院,北京100049

出　　处：《系统工程与电子技术》2024年第12期4231-4237,共7页Systems Engineering and Electronics

摘　　要：针对归一化最小和译码算法较置信传播译码算法误差较大的问题,提出自适应最小和译码算法。通过对当前迭代后验概率的硬判决值与前一次迭代后验概率的硬判决值进行计算,动态调整归一化因子与偏移因子,使得到的改进算法更接近于置信传播译码算法。在此基础上,应用分层式调度策略,提出分层自适应最小和译码算法,提升译码算法收敛速度。仿真实验结果表明,在误码率为10-6时,所提译码算法的误码性能与分层归一化最小和译码算法相比有0.25 dB的增益,与分层置信传播译码算法的译码性能十分接近,迭代次数仅有1次的增加,具有更好的收敛性能。To address the issue of larger errors in the normalized minimum sum decoding algorithm compared to the belief propagation decoding algorithm,an adaptive minimum sum decoding algorithm is proposed.By calculating the hard decision values of the current iteration posterior probability and the previous iteration posterior probability,the normalization factor and offset factor are dynamically adjusted to make the improved algorithm closer to the belief propagation decoding algorithm.Based on this,a hierarchical scheduling strategy is applied to propose a hierarchical adaptive minimum sum decoding algorithm,which improves the convergence speed of the decoding algorithm.Simulation results show that when the bit error rate is 10-6,the proposed decoding algorithm has a gain of 0.25 dB of bit error performance compared to the hierarchical normalized minimum sum decoding algorithm.Compared with the hierarchical belief propagation decoding algorithm,the proposed decoding performance has a similar performance,with only one increase in iteration times sum better convergence performance.

关 键 词：低密度奇偶校验码 分层自适应最小和译码算法 归一化因子 偏移因子 

分 类 号：TN911.22[电子电信—通信与信息系统]