Research And Optimization Of LT Codes Degree Distribution In Fountain Codes

There are generally two schemes for ensuring the reliability of data transmission in communication systems: one is to use a more stable communication protocol such as TCP/IP,and the other is to use Forward Error Correction(FEC).The TCP/IP protocol is not suitable for multicast and broadcast services in the communication field.Large-scale data distribution and reliable multicast/broadcast have become one of the problems to be solved urgently for multimedia services,and the digital fountain codes as a channel coding using forward error correction technology can solve this problem.LT codes(Luby Transform Codes)is a general-purpose digital fountain codes,and degree distribution is an important part of LT codes.The performance of the degree distribution directly affects the decoding success rate and bit error rate of the digital fountain codes,so the research on the degree distribution is of great significance to the digital fountain codes.This paper studies the optimization and design of the degree distribution of LT codes.First it describes the research background and significance of digital fountain codes,the current research status of LT codes at home and abroad,and the core technologies of digital fountain codes including: typical deletion channels applicable to digital fountain codes,encoding and decoding methods,and so on.Secondly,the principles of five classical degree distributions are discussed and four improvement schemes for degree distributions are given.The simulation results show that the four degree distributions given in this paper have different degrees of improvement in the decoding success rate compared to the four original classical degree distributions,and their decoding performance is superior.Thirdly,the concept of the chicken swarm optimization algorithm and the expected ripple size of degree distribution is explained,and two new joint degree distribution schemes are proposed: the single scale coefficient joint degree distribution and the double scale coefficient joint degree distribution.Finally,chicken swarm optimization algorithm,expected ripple size and Monte Carlo simulation are used to optimize the proposed joint degree distribution of single and double proportional coefficients.The simulation results show that the proposed two new joint degree distributions are compared with the classical degree distributions: The Ideal Soliton Distribution(ISD),Stationary Distribution(SD),Binary Exponential Distribution(BED)and Robust Soliton Distribution(RSD)had higher decoding success rate and lower bit error rate,so their decoding performance is superior.