论文部分内容阅读
基于原模图构造的低密度奇偶校验码(LDPC)性能很大程度上取决于扩展算法。为此,提出了一种构造准循环低密度奇偶校验码(QC-LDPC)的新方法。所述算法经过两步扩展得到QC-LDPC:第一步是原模图去重边,在边置换条件的约束下,使扩展所得矩阵局部围长最大化;第二步进行准循环扩展,通过计算机搜索得到规定长度内的所有闭环路径,比较环长和近似环路外信息度得到置换矩阵的最优偏移量,目的是剔除连通性差的短环对码性能的负面影响。对于不存在重边的原模图,则直接进行准循环扩展。仿真结果表明,利用该方法构造的QC-LDPC在译码门限和误码平层两方面都具有优异的性能。
Low-density parity-check codes (LDPC) based on the original model structure performance depends largely on the expansion algorithm. To this end, a new method for constructing quasi-cyclic low-density parity-check codes (QC-LDPC) is proposed. The algorithm obtains QC-LDPC in two steps: the first step is to deweight the edge of the original model and maximize the partial girth of the expanded matrix under the constraint of edge replacement. The second step is quasi-cyclic extension, Computer searches for all closed-loop paths within a specified length. The optimal offset of the permutation matrix is obtained by comparing the ring length and the approximate out-of-loop information. The purpose of this algorithm is to eliminate the negative impact of poorly-communicating short rings on code performance. For the original model without the heavy edge, the quasi-cyclic extension is performed directly. The simulation results show that the QC-LDPC constructed by this method has excellent performance in both decoding threshold and error floor.