The Encoding And Decoding Method In 5G Communication System

Polar code is one of the most important coding methods in 5G mobile communication systems.Polar code is formed by combining a series of independent channels to form a combined channel,and then split the combined channels to bit channels according to the channel transmission probability.The bit channels have the characteristic of polarization: a part of bit channels to be completely good channels,another part of the channels become completely noisy channels.Using this feature of the polar code,it is possible to transmit free information bits in perfectly good channels and transmit fixed bits those known by the both transmitted and received ends,such as check information in completely noisy channels.Through this method can ensure when the code length is substantial length,the channel capacity can reach Shannon limit.There are two key issues that need to be addressed: one is the structural problem of the polarization code,and the other is the problem of improving the decoding performance of the polarized code.Solving this problem is of great significance to the application of polarization codes in 5G channel coding.First of all,this paper introduces the development of communication system and channel coding process,and then expounds the theoretical basis of polarization code,including: channel polarization characteristic,the combination of the channel,channel separation and the basic framework of polarization code coding and decoding the basic process.Secondly,this paper studies the construction and decoding of polar code.In this paper,three classical polarization methods are considered in this paper: monte-carlo construction,gaussian approximation construction and density evolution.cosntuction In this paper shows the flow chart of the realization of these three methods,and analysis the performance of monte-carlo structure and gaussian approximation,it is concluded that the gaussian approximation decoding algorithm has better performance.On the polarization code decoding problem,in this paper,the successive cancellation algorithm(SC),list successive cancellation algorithm(LSC),cyclic check aid list successive algorithm(CRC-LSC)are studied.In three kinds of algorithm on the basis of the specific implementation process,the first SC decoding algorithm is a research on the relationship between decoding performance and the code length,then studied the LSC decoding algorithm of the decoding performance and the search width,the relationship between the three decoding algorithm is studied in the same conditions the decoding performance.Finally,this paper presents a system of segmented polarization coding.Through the analysis of the phenomenon of channel polarization and found the number of information bits need to transport is often greater than the number of complete polarization channel number,and decoding a bit of the mistakes are often those who are not transmit in the fully polarization code channel of.So this paper presents a segmented-LSC-CRC algorithm,by segment the information bits into subset information set to ensure all the free information bits are transmitted on the complete polarization channel.This paper presents the encoding and decoding methods of segmented-LSC-CRC.The simulation results show that the decoding performance is significantly improved,but on the sacrifice of channel utilization,so how to improve the utilization rate of channel is one of the focuses of future work.