Research On Decoding Algorithms Of Low-Density Parity-Check Code Based On Node Scheduling Strategy

With the continuous development of information technology,applications such as ultra-high-definition video,large-scale Internet of Things,unmanned driving,and cloud computing have emerged.These emerging applications put forward higher requirements for communication systems,such as larger bandwidth,higher transmission rate,lower delay,and higher system reliability.As one of the key technologies in modern communication system,channel coding directly affects the reliability of the communication system.The nonbinary low-density parity-check code(NB-LDPC),a channel coding scheme with performance approaching the limit of Shannon's theory,have gradually become the most competitive channel coding scheme today.To meet the challenge of lower complexity and lower latency for future communication system,this paper is devoted to the study of low-complexity decoding algorithms for NB-LDPC code.Our work starts with the method called the selected scheduling strategy,and the main contribution of this paper are listed below.1.In selected scheduling,the method of dividing node subset is the key step which affects the tradeoff between error correcting performance and computational complexity.To better balance the error correction performance and the complexity,this paper proposes a node-subset scheduling based on the stability of variable nodes(SNS).This algorithm judges whether the variable node is reliable through the metric of stability of variable nodes,and further divides the check nodes into the processing subset or the non-processing subset.Meanwhile,the judgment threshold is heuristically defined to match the iteration and variation of communication channel.We prove the proposed algorithm can effectively reduce the misjudgment which has a great impact on the error correction performance and avoid large performance degradations through an experiment.Finally,the results of many simulations show that the proposed algorithm can effectively reduce the complexity while maintaining good error correction performance compared to the original algorithm.2.Second,to further improve the error correction performance of the SNS while maintain low complexity,we studied the relevant laws of the decoder with the evolution of iterations after dividing the two types of node subsets.By designing three simulation experiments to study the number of elements in processing subset,the repetition ratio of the elements and the amplitude oscillation of the symbols in the extrinsic messages respectively.We discovered the “abnormal” stability in the error frames of the SNS-EMS algorithm and verified the correlation between this abnormal stability and the failure of the convergence of variable nodes.Based on the above experimental results,we propose a node subset scheduling algorithm with subset-reset mechanism.This algorithm uses a sliding window detection and a subset-reset mechanism to alleviate the performance degradation caused by the SNS scheduling algorithm.The elements in the processing subset will be reset,and all check nodes will update their messages in the next iteration after this abnormal stability have been detected.Then,the information transfer between the two types of nodes will be accelerate and the error correction performance of the SNS-EMS algorithm will be improved.The simulation results show that the proposed algorithm can reduce the complexity of the original M-EMS algorithm by approximately 25% without or nearly no degradation of error correction performance.