Study On Low-Complexity Spinal Code And Its Application

In order to meet the growing demand for communication technology in contemporary society,a lot of researchers have been attracted by a variety of coding and modulation techniques.spinal code is a new type of coding modulation technology,which combines the advantages of coding modulation,rateless code and random coding,and has a good throughput performance.In this thesis,we firtly propose a low complexity design scheme based on linear congruence method to replace the high-complexity Hash codes.Secondly,we propose a coding scheme to minimize the error symbol rate for spinal code in fading channel.In the first chapter,we mainly introduce the historical background and the state of the art of the coded modulation technologies,and give a brief introduction to the development of the rateless codes and the characteristics of the codes.In the second chapter,the theory of coding modulation technologies is described in detail.In this chapter,we first introduce the model of coding and modulation,and then introduce TCM,BICM,MLC and superposition coding,respectively.In the third chapter,the coding algorithm and the structure of the spinal codes are described in detail.One way to improve the performance of the spinal codes is adding tail symbols in the end of the code word.We give the theoretical analysis of the spinal codes with tail symbols based on its two-dimensional Euclidean distance spectrum.We also simulate the performance of the spinal codes with tail symbols on the complex Gaussian channel.The fourth chapter focuses on the low-complexity design of the spinal code.Firstly,we introduce the theory of the linear congruence method.Then to reduce the complexity of the Hash function,we propose a low-complexity pseudo-random generator based on the linear congruence method to replace the Hash function for the spinal codes.The proposed pseudo-random generator has almost the same structure as Hash function in the spinal codes.Finally we give the theoretical and statistical tests for the pseudo-random generator.We aslo give the complexity comparison between the proposed method and the Hash function.Chapter 5 focuses on the design of the spinal codes on fast fading channel.Firstly,the background of adaptive coded modulation technologies is briefly introduced.We firstly give an approximate calculation method of the symbol error rate based on mean of minimum Euclidean distance between symbols.We further propose a suboptimal resource allocation strategy to maximize the throughtput of the fading channel.The strategy is to minimize the square mean error of symbol error rate.We finally give the simulation results to exhibit the gain of the proposed resource allocation compared with the traditional method.Chapter 6 summarizes the thesis and gives the limitations of the research and the direction of further study.