Research On Polar Code Encoding And Decoding Algorithms

Polar Code is the first and,currently,the only family of codes with explicit construction to achieve the capacity when the length of code is infinite.The constrction of Polar Code is based on the channel polarization.At the same time, Polar Code has lower encoding and decoding complexity.Because of these characteristics,it has become one of hot spots on channel coding,and has a very good application prospect in the future.However,the channel can not polarized at finite length,so that the performance of Polar Code at finite length is poor.In this paper, the research purpose is improving the perfermence of Polar Code at finite length based on the successive decoding(SC).the main work of this paper is as follows:Firstly,deduce and simulate the performance of encoding and decoding of Polar Code.Production and properties of channel polarization phenomenon is derived,and simulate the polarization phenomenon.Deduce the formation of GN, information set selection and encoding.Summarize the method of information seclection.Analyze the SC decoding.At last,Conclude the reason of poor performance at finite length of Polar Code.Then, proposed CL-SCL decoding algorithm.The CL-SCL is based on the SCL. SCL can get a good performance at finite length,but it has high computational complexity.In this paper,using the threshold set partitioning position. When the channel capacity is good , reduce the length of list,it will reduce the computational complexity and get the same performance of SCL.At last, propsed a decoding of the interleaving cascade scheme.Comparative analysis of serial cascade and interleaving cascade. As to the interleaving cascade can't prevent the error transmitting, propose a new method of decoding. The method can stop error transmitting,improving the performance.?...