截断式原模图低密度奇偶校验卷积码边扩展优化

Edge Spreading Optimization for Terminated Protograph-based Low-Density Parity-Check Convolutional Codes

截断式原模图低密度奇偶校验(LDPC)卷积码(P-LDPC-CCs)结合了原模图LDPC(P-LDPC)码和卷积码的特点,具有多变的编码构造方式和优异的纠错性能,实现了编译码低时延特性。边扩展作为构造截断式原模图LDPC卷积码基础矩阵关键步骤,是影响其性能的重要因素。该文提出了一种边扩展优化方法。该方法利用原模图外信息转移(P-EXIT)算法理论分析基础矩阵的译码门限,引入差分进化思想搜索一定条件下最优的边扩展方式。理论分析与系统仿真结果均表明所提边扩展优化方法比现有的方法具有更好的性能。

Terminated Protograph-based Low-Density Parity-Check(LDPC)Convolutional Codes(Terminated P-LDPC-CCs),which combine the characteristics of Protograph-based LDPC(P-LDPC)codes and convolutional codes,have variable encoding constructed schemes,excellent error-correcting performance,and high-speed coding characteristics.As the key step of constructing Terminated P-LDPC-CCs,edge spreading is an important factor to determine the performance.In this paper,an edge spreading optimization method is proposed.In the proposed method,the differential evolution algorithm is introduced to search the best edge spreading mode based on the decoding threshold calculated by Protograph-based EXtrinsic Information Transfer(P-EXIT)analysis.Both P-EXIT analysis and simulation results indicate that the proposed edge spreading optimization method can achieve better performance.