Dissertation > Excellent graduate degree dissertation topics show

On the Efficient MAP Decoding, Non-data-aided SNR Estimation and Polynomial Interleaver Design for Turbo Coding Systems

Author: ZhaoHongYu
Tutor: FanPingZhi
School: Southwest Jiaotong University
Course: Communication and Information System
Keywords: Turbo Codes Decoding algorithm SNR estimation Interleaver Permutation polynomials
CLC: TN911.22
Type: PhD thesis
Year: 2011
Downloads: 137
Quote: 2
Read: Download Dissertation

Abstract

Turbo codes for channel coding transmission since the 1993 invention has excellent error correction performance close to the Shannon limit, will soon be applied to all kinds of communication system. Over the past 17 years, many scholars have published a lot of research on the theory and application of Turbo codes, covering simplified decoding algorithm decoding algorithm convergence analysis, excellent component codes, minimum code distance analysis and estimates interleaver design decoding algorithm parallel implementation. High efficiency based on the existing research results, the doctoral thesis focused on Turbo Codes-number field maximum a posteriori probability decoding number field value of the Log-MAP decoding algorithm standardized data-aided SNR estimation and replacement polynomial interleaver research. This paper first gives the Turbo code encoding and decoding principle, relevant research, analysis and literature review, we study the work of thinking, the main contribution of the paper and paper organizational structure \Next, from the classic standard MAP decoding algorithm is derived an improved MAP decoding algorithm. The algorithm does not introduce the logarithmic domain will be able to avoid the operator of logarithmic and exponential functions in the iterative process, so as to achieve higher coding efficiency, efficiency with similar efficiency and the Max-Log-MAP decoding algorithm. The improved MAP decoding algorithm for the error correction performance is optimal, suboptimal Max-Log-MAP algorithm will have to pay a consideration of the error correction performance. Subsequently, the paper proposes a two numerical standardized methods and pure integer arithmetic linear approximation Log-MAP algorithm. Standardization of these two numerical methods are called \Theoretical proof that the standardization of these two numerical methods are not interested in decoding performance impact. Addition standardized able to solve the problem of fixed-point overflow, and multiplication standardization can be applied based on pure integer arithmetic Turbo decoder or prove that Max-Log-MAP algorithm does not require any SNR estimation. In order to determine the appropriate multiplication standardized coefficient, the author gives two design methods of the probabilistic method and the equivalent length of the look-up table. Finally, given the simulation results of pure integer implementation of the 3GPP standard Turbo Decoding frame error rate (FER), 9-bit wide integer will be able to ensure optimal performance improved MAP algorithm is roughly equal to the error correction performance . Turbo code based on the received code block BPSK signal the moment estimation so as to realize the SNR estimation ideological In this paper, three non-data-aided SNR estimation were compared uncorrelated Rayleigh fading channel conditions. Three SNR estimation moment estimation method and the use of curve fitting method to calculate the SNR. The study found that the estimation error of curve fitting the deviation better able to compensate for the SNR of a shorter the Turbo code code block. The Turbo decoding simulation results show that the two proposed SNR estimation result is smaller than the existing literature SNR estimation error and to obtain a better bit error rate (BER) performance can be obtained, especially for the shorter length of the code block. Another advantage of these three SNR estimator do not need any information known SNR and Rayleigh fading parameter. Since 2005, on the ring of integers of quadratic permutation polynomial (QPP) is recommended to be used in the design of the Turbo code interleaver. The QPP interleaver does not require interleaver table and stored in the Turbo decoder having a uniformly distributed random interleaver is far superior performance. In addition, all QPP interleaver also has a maximum contention free characteristics, to avoid parallel Turbo decoder memory access violation. In this paper, permutation polynomial (PP) interleaver some new results, including: the structure ring of integers m times (m ≥ 1) PP's a simple method, quadratic and polynomial (QNP) proof of the necessary and sufficient condition QPP interleaving Exclusion equivalent count. Also search the excellent performance of the high-order permutation polynomial deinterleaver initial attempt, on the frame length N = 2048, code rate Rc = 1/3 of the 8-state Turbo code for a performance superior to the QPP interleaver and reach the famous the code matching interleaver (CMI) performance of high-order permutation polynomial interleaver.

Related Dissertations

CLC: > Industrial Technology > Radio electronics, telecommunications technology > Communicate > Communication theory > Information Theory > Channel coding theory
© 2012 www.DissertationTopic.Net Mobile