基于Tanner图边调度策略的BP算法  

BP algorithm based on Tanner graph edge scheduling strategy

作　　者：陈紫强[1] 侯田田 李亚云 

机构地区：[1]桂林电子科技大学无线宽带通信与信号处理省部重点实验室,广西桂林541004

出　　处：《电视技术》2017年第9期59-64,共6页Video Engineering

基　　金：国家自然科学基金项目(61461015;61371186;61261032;41201479);广西自然基金项目(2013GXNSFFA019004;2014jj AA70068);广西教育厅重点项目(ZD2014052)

摘　　要：为了降低LDPC码BP算法的计算复杂度,提高译码性能,提出基于边调度策略的BP(Edge-based Dynamic Scheduling for BP,EDBP)算法。经过几次迭代译码后,采用相邻节点中LLR信息比较小者更新校验方程节点的消息,降低计算复杂度,当检验点的可靠度达到某个阈值时,与之相邻的变量节点不再进行消息更新,避免了可靠度较低的变量节点LLR信息沿短环传播的影响,误码性能得到提高。仿真结果表明,在信噪比为3.0dB时,EDBP算法的计算复杂度仅为BP算法的42%,计算复杂度得到了降低,且EDBP算法的误码性能优于Flooding算法。In order to reduce the computational complexity and speed up the decoding of the BP algorithm for decoding Low-density parity check（LDPC） codes, The EDBP algorithm was proposed in this paper. After several iterations, using the part of the rela- tively smaller LLR information in adjacent nodes updates the message of the calibration equation, it reduces the computational com- plexity. When the check node reliability reaches the reliability threshold, the adjacent variable nodes no longer update the mes- sage. It avoids the influence of low reliability variable nodes LLR information propagating along short loop. So the bit-error-rate performance is improved. The simulation results show that when the SNR is 3.0dB, the EDBP algorithm computational complexity is only about 42% of the BP algorithm complexity, the computational complexity is reduced and the performance of the proposed decoding algorithm is better than Flooding decoding algorithm.

关 键 词：LDPC码 TANNER图 误码性能 

分 类 号：TN911.22[电子电信—通信与信息系统]