摘要
长LDPC码的Tanner图中通常没有环路,此时LLR BP算法是性能最优的软判决译码算法。而短LDPC码的Tanner图中通常存在环路,因此变量节点之间的信息就不再相互独立,这时LLR BP算法的译码性能就会下降。提出一种改进型LLR BP算法,利用概率论和最小均方误差准则来计算该算法中的参数。仿真结果表明改进型LLR BP算法比LLR BP译码算法、Normalized BP算法以及Offset BP算法具有更好的LDPC译码性能。
The Tanner graph of long LDPC codes usually has no loops and LLR BP algorithm is the best soft-decision decoding algorithm.But the Tanner graph of short LDPC codes usually has loops,so the information within variable nodes isn't mutually independent and the decoding performance of LLR BP algorithm will decrease.A modified LLR BP algorithm is proposed.Parameters of the proposed algorithm are calculated according to probability theory and minimum mean square error rule.The simulation results show that the LDPC decoding performance of modified LLR BP algorithm is more effective than that of LLR BP algorithm,Normalized BP algorithm and Offset BP algorithm.
出处
《武汉理工大学学报》
CAS
CSCD
北大核心
2010年第10期174-179,共6页
Journal of Wuhan University of Technology