Research On Decoding Algorithm Of Non Binary LDPC Code

In order to ensure the high quality transmission data in communication system,channel coding technology occupies very important position,in a number of channel coding scheme,the binary low-density parity check(NB-LDPC)code is a kind of channel coding scheme,of which the error correction performance is closer to the Shannon coding limit than binary LDPC codes.Although a variety of decoding algorithms have been developed,in practical application,the contradiction between decoding performance and complexity is still unresolved.In general,if the decoding algorithm performs better in decoding,its complexity will be higher,but if the complexity is reduced,the decoding performance will be worse.In this paper,two decoding algorithms of non-binary LDPC codes are studied.The main contents include:(1)Targeting at the poor decoding performance of iterative hard reliability decoding algorithm,a reliability updating method is designed to expand the range of alternative codes.The key point is that both the elements with a Hamming distance of zero from checksums and those of 1 from checksums are updated when the reliability is updated,which expands the range of alternative code elements,increases the probability of decoding success and improves the bit error rate performance.Simulation results show that the codes error rate performance of the improved iterative hard reliability decoding algorithm can be improved by about 0.2dB.Since the weighted iterative hard reliability decoding algorithm has the same principle as the iterative hard reliability decoding algorithm,therefore,the reliability updating method to expand the range of alternative codes is adjusted and applied to weighted iterative hard reliability decoding algorithm,designing a weighted iterative hard reliability decoding algorithm based on expanding range of alternative codes to improve decoding performance.Simulation results show that compared with the original algorithm,the codes error rate performance can be improved by about 0.1dB.(2)In order to improve the decoding performance of the two extra column trellis min max decoding algorithm,a new method of calculating the confidence information of variable nodes is designed.The main idea of the improved scheme focuses on decreasing errors from confidence calculation to improve decoding performance by using threshold value and correction factor.Simulation results indicate that the performance of the improved decoding algorithm can be improved by about 0.2dB.