用于实现并行行列联合译码的QC-LDPC码构造方法  

Quasi-cyclic LDPC code construction for implementation of parallel joint row-column decoding

作　　者：董明科[1] 王达[1] 郑雅丹[1] 项海格[1] 

机构地区：[1]北京大学信息科学技术学院、区域光纤通信网与新型光通信系统国家重点实验室,北京100871

出　　处：《高技术通讯》2012年第6期559-565,共7页Chinese High Technology Letters

基　　金：国家自然科学基金（61071083,60972008）和国防基金（9140A220310）资助项目.

摘　　要：针对普通低密度校验（LDPC）码制约行列联合（JRC）译码算法并行度提高的问题，基于块渐进边增长（BPEG）算法，提出了一种用于并行JRC译码的LDPC码构造方法。该方法构造的准循环LDPC码（QC-LDPC）基矩阵由含r（r为大于1的整数）行的行组构成，允许一个行组内的r 行进行并行JRC运算。仿真结果表明，用上述构造方法构造的LDPC码与BPEG码的误码性能相当。硬件实现表明，用此构造码的并行译码器的速率能达到典型传统准循环译码器的3倍以上，为面向译码器的LDPC码构造提供了范例。Aiming at the restriction of normal low-density parity check （LDPC） codes on the parallelism lift of joint row- column （JRC） decoding algorithms, this paper proposes a new LDPC code construction method for parallel JRC de- coding based on an improved block progressive edge-growth （BPEG） algorithm. When using this method, the rows of the quasi-cyclic LDPC （QC-LDPC） base parity check matrix are organized into groups of r rows. Parallel JRC decoding can be performed within each group. The simulation result shows that the performance of the LDPC codes constructed by the proposed method is similar to that by the BPEG method. However, the hardware implementation on the codes constructed by the proposed method shows that the throughput of the parallel decoder is more than three times of that of the typical traditional QC-LDPC decoders. This provides an example of decoder-oriented LD- PC code design.

关 键 词：低密度校验(LDPC)码 行列联合(JRC)译码 渐进边增长(PEG)算法 块渐进 边增长(BPEG)算法 准循环LDPC(QC—LDPC)码 面向译码器的LDPC码构造方法 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]