Research On Decoding Algorithm Of Spatially Coupled LDPC Codes Based On Protograph

With the advancement of science and technology and the improvement of living standards,people's demand for communications and the quality of communications have also increased.Error correction codes are an important means to improve the reliability of information transmission.Low-Density Parity-Check(LDPC),as a type of error correction code,has attracted much attention due to its excellent performance.Among them,the spatially-coupled low-density parity-check(SC-LDPC)code's decoding limit performance is closer to the Shannon limit,providing a theoretical basis for high reliability of information transmission.For the decoding of SC-LDPC codes,the confidence decoding algorithm needs to wait until the entire code length sequence is received before starting.When the code length is large,the decoding requires a large amount of memory and causes high delay.In view of this problem,based on the special structure of SC-LDPC codes,Liva proposed a solution for window decoding.The advantages of window decoding are low latency and low complexity.Under the condition that the signal-to-noise ratio is constant,conventional Belief-Propagation(BP)decoding is run in the window.The bit error rate decreases as the width of the window increases,and the performance of the window decoding reaches the traditional Large-scale Pipeline decoding effect.The window decoding obtains low latency and low complexity at the cost of lossy decoding performance.How to reduce the loss of decoding performance as much as possible is the research focus of this paper.First,this article introduces the research background of SC-LDPC codes,the representation and construction of SC-LDPC codes based on the original model diagrams,and systematically introduces the principles of window decoding algorithms for SC-LDPC codes,and compares them through computer simulation.The effect of different window sizes on decoding performance.Secondly,this paper introduces a design scheme for window decoding of SC-LDPC codes with higher decoding performance.This scheme proposes to effectively terminate the decoding and reuse the edge message of the previous target symbol.Compared with the traditional window decoding,it also improves its performance gain,reduces its complexity and delay.In order to further improve the decoding performance,based on the original design scheme,this paper proposes an improved method of combining early termination decoding and dynamic adjustment of the window size for window decoding of SC-LDPC codes.This method monitors the dynamic change of the window size and the average number of iterations of the corresponding window.Through simulation analysis under the additive Gaussian White Noise(AWGN)channel,compared with traditional window decoding,the bit error rate is reduced.Finally,this paper attempts to concatenate the LLR BP algorithm and the residual-based confidence propagation decoding algorithm into the window decoding algorithm,and seek the best decoding scheme by appropriately changing the length of the target symbol in the window.