适合行列合并译码算法的LDPC码的构造  

Construction of LDPC codes suitable for the row-column joint decoding algorithm

作　　者：贺一峰[1] 段哲民[1] 

机构地区：[1]西北工业大学电子信息学院,西安710072

出　　处：《高技术通讯》2015年第7期677-684,共8页Chinese High Technology Letters

基　　金：国家自然科学基金(61071083);973计划(2009CB320403)资助项目

摘　　要：针对低密度奇偶校验(LDPC)码译码器使用行列合并(RCJ)译码算法时存在的流水线冲突问题,提出了一种构造适合行列合并译码算法的LDPC码的方法。该构造方法在渐进添边(PEG)构造算法的基础上,加入行运算顺序的约束条件,使得利用该算法构造的LDPC码在行运算顺序上相邻两行内的非零块不同时共用一列,从而避免了硬件布局布线冲突,减少了流水线延时,提高了译码速率。仿真结果表明,用这种方法构造的LDPC码与全球微波互联接入(WiMAX)标准给出的LDPC码的译码性能相当,而且有效地减少了译码迭代次数,降低了硬件实现复杂度,提高了LDPC译码器吞吐率。To deal with the pipeline collision problem of a low density parity check （LDPC） codes＇ decoder using the row-column joint （RCJ） decoding algorithm, a method for construction of the LDPC codes suitable for the RCJ decoding algorithm is proposed. The proposed construction method adds a constraint condition in the order of line operation on the basis of the progressive edge-growth （PEG） construction algorithm, to guarantee that the non-zero matrix in two adjacent lines does not share the same column in the order of line operation to avoid the pipeline collisio n, reduce the pipeline delay the LDPC codes constructed by the communication standard the number of decoding iterati throughput of LDPC decoders. , and increase the decoding rate. The simulation results show that the performance by using the proposed method is about the same as that of the LDPC codes proposed of WiMAX. Meanwhile, the proposed construction method can effectively decrease on, reduce the complexity of decoder implementation, and greatly improve the

关 键 词：高速通信 LDPC码 行列合并译码算法 PEG构造算法 流水线冲突 

分 类 号：TN911.22[电子电信—通信与信息系统]