提示
低复杂度marker-ldpc码级联方案研究与实现
首发时间：2025-03-10
摘要：针对基于加权莱文史特距离的级联码译码复杂度高的问题，采用归一化最小和（normalizedmin-sum，nms）算法作为级联码的外译码算法，在降低复杂度的同时，保证性能逼近传统译码方案。进一步，提出硬判决迭代译码算法，将外译码器的输出反馈给内译码器，更新前向及后向度量值，提升级联码纠正插入/删节错误的能力。仿真结果表明，基于nms算法的级联方案在性能损失8%的前提下降低了约34%外译码复杂度。当误帧率为0.02且替代错误概率为0.001时，硬判决迭代译码方案比传统的非迭代译码方案能多纠正8个插入/删节错误。
关键词：
for information in english, please click here
research and implementation of a low-complexity marker-ldpc code concatenation scheme
abstract：to address the high decoding complexity of concatenated codes based on weighted levenshtein distance, the normalized min-sum (nms) algorithm is adopted as the outer decoder for concatenated codes. this approach reduces complexity while ensuring performance remains close to that of traditional decoding schemes. furthermore, a hard-decision iterative decoding algorithm is proposed, where the output of the outer decoder is fed back to the inner decoder to update forward and backward metrics, enhancing the capability of concatenated codes to correct insertion/deletion errors. simulation results show that the concatenated scheme based on the nms algorithm reduces outer decoding complexity by approximately 34% at the cost of an 8% performance loss. when the frame error rate is 0.02 and the substitution error probability is 0.001, the hard-decision iterative decoding scheme can correct up to 8 more insertion/deletion errors compared to the traditional non-iterative decoding scheme.
keywords：
基金：
论文图表：
引用
导出参考文献
no.****
动态公开评议
共计0人参与
勘误表
低复杂度marker-ldpc码级联方案研究与实现
评论
全部评论