面向5G通信系统的极化码编译码研究

发布时间：2018-04-08 11:56

本文选题：极化码　切入点：极化码构造　出处：《深圳大学》2017年硕士论文

【摘要】：信道编码是5G移动通信中一种重要的手段,极化码是其中一种编码方式。通过将一系列相同独立的信道进行组合,再把组合信道进行拆分后形成位信道。形成的位信道具有了极化的特性,也就是说在这些位信道中一部分容量等于1,另外一部分容量为0。利用极化码的这个特性,可以在完全好的信道中传输自由信息比特,在完全噪声的信道中传输固定比特,也就是发送端和接收端都知道的信息,例如:校验信息。通过这种方法可以保证在码长趋于无限长的时候,使信道的容量趋于香农限。极化码有两个关键问题需要解决:一个是极化码的构造问题,另外一个是提升极化码的译码性能的问题。解决这个问题对于极化码在5G信道编码中的运用具有重要的意义。首先,本文介绍通信系统与信道编码的发展过程,接着对极化码的理论基础进行了阐述,包括:信道极化的特性、信道的组合、信道的拆分以及极化码编码的基本框架和译码的基本流程。其次,本文针对极化码的构造和译码问题进行了研究。在极化码的构造上本文考虑三种经典的极化码构造方法:蒙特卡罗构造、高斯近似构造和密度进化构造。在文中给出这三种方法的实现流程图,并且针对蒙特卡罗构造和高斯近似构造进行了性能分析,得出了高斯近似译码算法具有更好性能的结论。在极化码的译码问题上,本文对接续取消译码算法(SC)、列表继续取消译码算法(LSC)、和有循环校验的列表接续取消算法(CRC-LSC)进行了研究。在给出三种算法具体实施过程的基础上,首先SC译码算法研究其译码性能与码长的关系,随后研究LSC译码算法那研究了译码性能与搜索宽度的关系,最后研究这三种译码算法在相同条件下的译码性能。最后,本文提出一种分段极化码编译码系统。通过分析极化码信道极化的现象,发现需要传输的信息比的数目往往大于完全极化的信道数目,并且在译码时出现的错误比特往往是那些在没有完全极化码的信道中传输的比特。因此本文还提出一种分段极化码编译系统,通过将需要发送的信息进行合理的分段,从而保证所有的自由信息比特都在完全极化了的信道中传输,并且给出了在此分段系统中的译码性能。本文给出该系统编码和译码的具体实施方法。仿真结果显示,译码性能得到了显著提高,但是在牺牲信道的利用率上得到的,因此如何提高信道的利用率是未来工作的一个重点。
[Abstract]:Channel coding is an important means in 5 G mobile communication, and polarization code is one of the coding methods.By combining a series of identical independent channels, the combined channels are split into bit channels.The resulting bit channels have polarization characteristics, that is, one part of the bit channel capacity is equal to 1, the other part of the capacity is zero.Using this property of polarization codes, free information bits can be transmitted in perfect good channels and fixed bits in fully noisy channels, that is, information known to both sender and receiver, such as check information.This method can ensure that the channel capacity tends to Shannon limit when the code length tends to be infinite.There are two key problems to be solved in polarization codes: one is the construction of polarization codes and the other is to improve the decoding performance of polarization codes.Solving this problem is of great significance for the application of polarization codes in 5G channel coding.First of all, this paper introduces the development of communication system and channel coding, and then describes the theoretical basis of polarization code, including: the characteristics of channel polarization, channel combination,The channel splitting, the basic frame of polarization code coding and the basic process of decoding.Secondly, the construction and decoding of polarization codes are studied in this paper.In this paper, we consider three classical construction methods of polarization codes: Monte Carlo construction, Gao Si approximation structure and density evolution structure.In this paper, the flow chart of these three methods is given, and the performance analysis of Monte Carlo structure and Gao Si approximate structure is carried out, and the conclusion that Gao Si approximate decoding algorithm has better performance is obtained.On the problem of decoding polarization codes, this paper studies the continuation cancellation decoding algorithm, the list continuation cancellation decoding algorithm, and the cyclic check list continuation cancellation algorithm, CRC-LSC. in this paper, we study the following algorithms: the continuation cancellation algorithm, the list continuation cancellation algorithm, and the list continuation cancellation algorithm (CRC-LSCC).Based on the detailed implementation of the three algorithms, the relationship between decoding performance and code length of SC decoding algorithm is studied, and the relationship between decoding performance and search width of LSC decoding algorithm is studied.Finally, the decoding performance of the three decoding algorithms under the same conditions is studied.Finally, a piecewise polarization codec system is proposed.By analyzing the polarization phenomenon of polarization codes, it is found that the number of information ratios to be transmitted is often greater than that of fully polarized channels.And the error bits that occur in decoding are those that are transmitted in channels that do not have fully polarized codes.Therefore, a piecewise polarization codec system is proposed to ensure that all free information bits are transmitted in a fully polarized channel by reasonably segmenting the information that needs to be sent.The decoding performance in this piecewise system is also given.In this paper, the coding and decoding methods of the system are given.Simulation results show that the decoding performance has been significantly improved, but achieved in sacrificing the channel utilization, so how to improve the channel utilization is one of the key tasks in the future.
【学位授予单位】：深圳大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN911.22;TN929.5

【相似文献】