Research On Complexity Optimization Of Non-binary LDPC Encoding And Decoding Algorithm

In the wireless channel coding and decoding code,the non-binary Low-Density Parity-Check(LDPC)code has a more obvious advantage than the binary LDPC code,and has a higher coding gain.But at the same time increase the complexity of the encoding and decoding algorithm,which is not conducive to application in hardware.Therefore,reducing the complexity of the encoding and decoding algorithm while ensuring higher coding gain,so as to facilitate the implementation of the hardware platform,is currently the main problem facing the application of non-binary LDPC code engineering.This paper optimizes the three aspects of check matrix construction algorithm,coding algorithm and decoding algorithm in non-binary LDPC code,and constructs non-binary LDPC code with medium and short code lengths with strong error correction performance,which has important theoretical and engineering application value.The thesis is devoted to an effective trade-off between decoding performance and computational complexity,and studies the multivariate LDPC code algorithm which can equalize system complexity and error correction performance.The main research contents are as follows:Firstly,the basic definition of multivariate LDPC codes and the way of three-stage algorithm performance are given.The principle of coding algorithm and decoding algorithm is theoretically deduced,which lays a foundation for the optimization research of check matrix construction algorithm and coding algorithm.Then,the binary iterative iterative coding algorithm is improved to reduce the coding complexity.An iterative coding algorithm suitable for non-binary codes is proposed.In order to reduce the high coding complexity,the constructed non-binary LDPC code check matrix can be applied to the non-binary iterative coding algorithm with linear coding complexity.A hybrid check matrix construction algorithm is proposed.The algorithm mainly improves the check matrix construction algorithm from two aspects of coding algorithm and check matrix construction,which effectively reduced the coding complexity.The base matrix in the check matrix is constructed by the Progress Edge Growth(PEG)algorithm with the lower triangular structure,and applied to the improved quasi-cyclic LDPC(Quasi-Cyslic,QC)code.The multi-LDPC code with excellent error performance and low coding complexity provides a theoretical reference for implementation on the hardware platform.Finally,research is carried out from the soft decision decoding algorithm and hard decision decoding algorithm.From the perspective of the soft decision decoding algorithm,the complexity analysis and simulation analysis of various soft decision decoding algorithms are performed from the frequency domain and Log domain respectively,and choose a soft decision decoding algorithm that is more suitable for wireless mobile communication.From the perspective of the hard decision decoding algorithm,the error correction performance of the hard decision decoding algorithm is poor,but the decoding complexity is low.Therefore,the error correction performance is improved,and an sum of magnitude for hard decision decoding algorithm based on loop update detection is proposed.This algorithm can speed up the convergence speed of decoding and enhance system reliability,while ensuring no increase in decoding complexity.In case of improving the performance of the hard decision decoding algorithm,it is a good alternative decoding algorithm from the perspective of the wireless mobile communication system.