标准低密度奇偶校验码译码算法中量化结构  

Design of quantitative structure of LDPC decoded algorithm

作　　者：兰亚柱[1,2] 杨海钢[1] 林郁[1] 

机构地区：[1]中国科学院电子学研究所可编程芯片与系统研究室,北京100190 [2]中国科学院大学,北京100190

出　　处：《太赫兹科学与电子信息学报》2015年第4期584-589,共6页Journal of Terahertz Science and Electronic Information Technology

基　　金：国家自然科学基金资助项目(61404140;61271149;61106033)

摘　　要：DVB-S2标准低密度奇偶校验码(LDPC)译码器在深空通信中面临着低复杂度、高灵活性及普适性方面的迫切需求。通过对LDPC译码算法中量化结构的研究,提出一种动态自适应量化结构的设计方法。该方法在常规均匀硬件量化的基础上,提出了修正化Min-Sum译码算法中的数据信息初始化及迭代译码的动态自适应量化结构,解决了DVB-S2标准LDPC码译码时存在的校验节点运算与变量节点运算之间的复杂度不平衡的问题,并由此提高了译码器的译码性能。实验证明,以DVB-S2标准LDPC码中码长为16 200,码率为1/2的为例,提供动态自适应量化结构与常规的均匀量化结构相比,节省硬件资源为4%。此外,动态自适应量化结构支持动态可配置功能,保证了DVB-S2标准LDPC译码器的灵活性及普适性。In deep space communications, the Low Density Parity Check(LDPC) Codes decoding based on DVB-S2(Second Generation Satellite Digital Video Broadcasting Standard) must meet the requirements of low complexity, high flexibility and universal aspects. This paper presents a methodology to design a dynamic adaptive quantitative structure based on studying the quantitative structure of LDPC decoding. Based on conventional uniform quantization of the hardware, a dynamic adaptive quantization structure for data information initialization and iteration decoding of modified Min-Sum decoding algorithm is proposed, which reduces the complexity imbalance between the check node processing units and variable node processing units, therefore, improves the decoding performance of the decoder. Experiments show that, for the DVB-S2 standard LDPC code with code length of 16 200 and rate of 1/2, the proposed dynamic adaptive quantization structure saves 4% hardware resources compared with the conventional adaptive quantization structure. In addition, the dynamic adaptive quantization structure supports dynamic configuration functions, which ensures the flexibility and universality of LDPC decoder.

关 键 词：DVB-S2标准 低密度奇偶校验码 译码器优化设计 量化结构 

分 类 号：TN911.22[电子电信—通信与信息系统]