Turbo码编解码的算法研究及FPGA设计与实现

发布时间：2018-02-27 20:49

  本文关键词： Turbo编译码 MAX-Log-MAP 线性拟合 比例因子 FPGA　出处：《武汉理工大学》2015年硕士论文　论文类型：学位论文

【摘要】：Turbo码以其具有良好的逼近香农极限的译码性能,在无线通信领域中起到了举足轻重的作用,并成了当今信息时代研究纠错编码的重要课题之一。在人们生活质量不断提高的信息时代,通信系统中对工具的便捷性和技术的有效性要求也越来越严格,以促使人们对通信系统的关注越来越紧密。现在随着4G的推出,人们能够享受4G带来的更快的速度和更高的性能,而这一进步正是由于通信工程师们对通信协议、算法等不断的研究和改进的成果,其中,Turbo码作为通信系统中常用的编解码算法之一,在算法的研究上也不断的得到了改进和创新。通过研究Turbo码编码器和译码器结构,并详细介绍了编解码各个模块的实现原理。其中,对于编码模块,详细分析了分量编码器和交织器两个关键问题;对于译码模块,对比了Turbo码常用的两种译码算法:SOVA和MAP,并详细针对MAP算法进行了数据推导和分析,同时针对码率、交织长度和迭代次数的不同分别研究了Turbo码的译码性能。本文对MAX-Log-MAP算法主要从以下两个方面做了改进,第一方面,对MAP算法两路信号局限性进行了扩展分析,当有多路信号时,对算法进行了优化和近似分析;另一方面,针对MAX-Log-MAP算法中忽略的修正函数入手,考虑了修正函数的影响,并对修正函数进行了线性拟合,同时加入了外信息的比例因子,并且从性能和复杂度角度分析算法的有效性,通过MATLAB工具对算法的性能从不同的角度进行了验证。得到了以下结论:在损失了一定复杂度的前提下,有效地提高了算法的译码性能。在FPGA的实现中,借助XILINX设计软件ISE12.2,完成了对编码码率为1/3,量化宽度为7bits的Turbo码编解码结构的方案设计,并通过ISE软件仿真平台使用硬件编程语言Verilog,完成了编解码的FPGA实现,在编码实现结构中,完成了分量码、交织、并串和串并的转换,借助了三分频的思想。在信道端,由于信号在实际的传输过程中会受到噪声的干扰,为了模拟真实的信号通信过程,所以本文中还借助MATLAB模拟了离散的AWGN信道。在译码实现结构中,分别采用MAX-Log-MAP算法和新提出的改进算法的译码器完成译码功能,并对仿真结果进行验证。根据对Turbo码性能的分析,以及Turbo码在各个领域的广泛应用,其算法和结构会不断的得到研究和改进,进而成为通信行业中最有效的编码技术。
[Abstract]:Turbo codes play an important role in wireless communication field because of their good decoding performance of approaching Shannon limit. And it has become one of the important topics of error correction coding in the information age. In the information age where people's quality of life is improving, the convenience of tools and the effectiveness of technology are becoming more and more strict in communication system. Now with 4G, people can enjoy the faster speed and higher performance that 4G brings, and this progress is due to the communication protocols that communications engineers have. As one of the commonly used coding and decoding algorithms in communication systems, turbo codes have been continuously improved and innovated in the research of the algorithms. The realization principle of each module is introduced in detail. The key problems of component encoder and interleaver are analyzed in detail for the coding module, and for the decoding module, In this paper, two common decoding algorithms of Turbo codes, namely: SOVA and MAPP, are compared, and the data of MAP algorithm is deduced and analyzed in detail, and the code rate is also analyzed. In this paper, the decoding performance of Turbo codes is studied with different interleaving lengths and iterations. The MAX-Log-MAP algorithm is improved from the following two aspects. Firstly, the limitations of the two channels of MAP algorithm are extended and analyzed, when there are multichannel signals, The algorithm is optimized and approximate analyzed. On the other hand, the influence of the correction function is considered, and the correction function is fitted linearly, and the proportion factor of the external information is added to the modified function, which is ignored in the MAX-Log-MAP algorithm. The effectiveness of the algorithm is analyzed from the point of view of performance and complexity. The performance of the algorithm is verified from different angles through MATLAB tools. The following conclusions are obtained: under the premise of losing a certain complexity, In the implementation of FPGA, with the help of XILINX design software ISE12.2, the scheme of coding and decoding Turbo codes with 1 / 3 coding rate and 7 bits quantization width is completed. Through the ISE software simulation platform, using the hardware programming language Verilog, the FPGA implementation of coding and decoding is completed. In the coding implementation structure, the conversion of component code, interleaving, parallel string and series-parallel is completed, and the idea of three-frequency division is used. In order to simulate the real signal communication process, the discrete AWGN channel is simulated by MATLAB in order to simulate the real signal communication process because the signal will be disturbed by noise in the actual transmission process. The decoder of the MAX-Log-MAP algorithm and the new improved algorithm are used to complete the decoding function, and the simulation results are verified. According to the analysis of the performance of the Turbo code and the wide application of the Turbo code in various fields, Its algorithm and structure will be continuously studied and improved, and then become the most effective coding technology in the communication industry.
【学位授予单位】：武汉理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TN791;TN911.22

【参考文献】

中国期刊全文数据库 前5条

1 毕岗;陈国宏;王建毅;;高性能的Max-Log-MAP线性分段算法研究[J];电路与系统学报;2012年06期

2 张吾进;毛倩;;基于Turbo码的纠错加密联合编码方案[J];无线电通信技术;2012年05期

3 王宁;陈名松;杜晓萍;;Turbo码的研究及仿真[J];通信技术;2012年03期

4 吴丹;唐秋菊;梁保卫;;删余Turbo乘积码的编译码算法分析[J];无线电工程;2011年12期

5 刘洋;;统一架构FPGA使功能设计收放自如[J];电子设计技术;2010年09期

中国博士学位论文全文数据库 前3条

1 莫磊;无线传感器/执行器网络协同感知与控制研究[D];华南理工大学;2013年

2 王颖;重叠复用技术应用在OFDM系统及其Turbo码技术的研究[D];北京邮电大学;2013年

3 陈晓刚;现代编码的性能分析与简化译码算法[D];北京邮电大学;2010年

中国硕士学位论文全文数据库 前8条

1 甫榆;高吞吐率Turbo译码器设计与实现[D];西南交通大学;2014年

2 丁旭辉;低信噪比环境下基于低码率Turbo码的可靠通信传输技术研究[D];西南交通大学;2014年

3 何贤国;Turbo码中伪随机交织器的设计与优化[D];杭州电子科技大学;2014年

4 邬可;Turbo码与网络编码的联合设计在协作通信中的应用研究[D];南京邮电大学;2013年

5 赵晓刚;LTE系统中信道编码的研究及Turbo码的FPGA实现[D];西安电子科技大学;2013年

6 胡嫦春;Turbo码在LTE系统中的应用及FPGA实现[D];武汉理工大学;2012年

7 顾超;Turbo码与LDPC码编译码的研究[D];南京航空航天大学;2012年

8 李超;TD-SCDMA系统中Turbo码的研究与FPGA实现[D];中南大学;2011年

，

本文编号：1544245