Optimization And Design Of Low-Density Parity-Check Code Decoding Algorithm

With the continuous development of communication technology,LDPC codes have been widely used in many fields such as wireless communication,wired communication,satellite communication,and data storage.As a coding scheme that has been extensively studied in recent years,LDPC codes have attracted much attention due to their excellent error correction performance and low-complexity decoding algorithms.The performance of LDPC codes and the efficiency of decoding algorithms directly affect the reliability and performance of communication systems.Therefore,the purpose and significance of researching LDPC decoding algorithms is to continuously improve their error correction performance and convergence speed to meet the demand for high-speed,high-capacity and low BER communication system.Based on the above requirements,this paper focuses on the optimization and design of the traditional EMS algorithm and RBP algorithm,starting from the decoding performance and convergence speed of the decoding algorithm,etc.The main research contents and work are summarized as follows:1.The two aspects of EMS algorithm node update order and external information value oscillation are firstly considered.The difference of node update order affects the convergence speed and performance of decoding.To further optimize the EMS algorithm,this paper introduces a mix-and-wash scheduling strategy to update the nodes of the algorithm by column to ensure that the check nodes can get the latest variable node information at each round of iteration.In addition,to mitigate the effect of information oscillation,the variable node update is improved in this paper.By introducing a weighting factor γ = 0.9,a weighted average of the information values before and after two iterations is applied to smooth the fluctuations of the variable node information values.Finally,this paper combines the two previous improvement methods to propose an EMS algorithm based on mixed wash scheduling strategy and variable node update improvement.The experimental simulation results show that the Shuffled-VNU-EMS algorithm designed in this paper improves the coding gain of 0.2～0.4d B to the original algorithm with a slight increase in decoding complexity,and improves the convergence speed of the EMS decoding algorithm by about 24% at maximum.2.Then,this thesis optimizes and designs the RBP algorithm that adopts a dynamic scheduling strategy.Considering that the RBP algorithm only updates the least reliable variable nodes without considering whether the messages used for updating are reliable,this paper proposes a reliability measure based on the stability of variable nodes(BNS-RBP)from the reliability of variable nodes.By finding the most stable variable node adjacent to the least reliable node,the method proposed in this paper can improve the decoding performance by screening out reliable messages for update propagation.Then,in order to further alleviate the greediness of the decoding algorithm,a decay factor μ=0.9 is introduced on the basis of the BNS-RBP algorithm,and a decay mechanism to prevent resource wastage is proposed to decay the variable nodes that are selected as the most unreliable multiple times to ensure that other nodes can also be updated,thus improving the error correction performance of the decoding algorithm.The results of simulation tests show that the proposed algorithm in this paper has better error correction performance and faster convergence speed than the RBP algorithm and the NW-RBP algorithm.