融合DFT及小波去噪的OFDM信道估计算法研究

发布时间：2018-01-07 19:07

本文关键词：融合DFT及小波去噪的OFDM信道估计算法研究　出处：《江西理工大学》2017年硕士论文　论文类型：学位论文

更多相关文章： OFDM 信道估计 DFT 阈值 小波去噪

【摘要】：随着科技的进步,越来越多的新技术被提出以及被使用,OFDM技术作为LTE的关键技术成为了众多研究人员研究的热点。运用OFDM技术的目的是防止符号间的干扰以及信道之间的干扰。作为OFDM技术关键部分的信道估计技术得到了重点关注,信道估计出现使得恢复出的信道状态信息变得更加精确,使系统性能得到了提升。本文围绕OFDM估计方法进行深入探究,对于经典估计方法不足地方提出相应地改进算法,对于研究内容以及创新点进行如下阐述。(1)研究了无线信道基本特性,分析了在实际环境当中无线信道经历的各种衰落。对于信号在无线信道之中受到了符号间干扰以及信道间干扰情况,研究OFDM原理以及其关键技术,其中介绍了OFDM导频的三种结构以及三种常见估计算法。针对如何能够得出良好信道响应值的问题探究了几种实际运用当中常见的插值算法。(2)研究了在阈值基础上面进行改进DFT估计方法,信号经过离散化之后往往会造成频谱泄露,针对信道当中出现的这种情况提出了加入海明窗进行解决。对于经典DFT估计算法不能够滤除掉循环前缀以内噪声的问题,提出了加入改进阈值进行去噪的方法。该阈值法能够很好地消除信道中突发大脉冲噪声的影响,并对该算法进行MATLAB仿真实验。实验结果显示,改进算法在误比特率方面最小以及在均方误差方面也是最小,所以该改进算法能够很好地提升估计性能。(3)研究了在小波基础上面进行改进DFT估计方法,对于传统的DFT-MMSE估计算法不足提出加入小波改进方法。相对于DFT与阈值相融合算法不足点还提出了运用小波去噪以及阈值去噪相融合的改进算法。该算法是在LS信道估计之后加入小波分解并通过运用软阈值法对信号进行去除噪声操作,紧接着对得出的信号进行离散傅里叶反变换后再一次进行阈值处理来更深程度地减少噪声。通过仿真实验显示,这两种算法能够很好地减少噪声影响,提升了系统的性能。
[Abstract]:With the progress of science and technology, more and more new technologies have been proposed and used OFDM technology as the key technology of LTE has become the research focus of many researchers. The use of OFDM technology to prevent interference between inter symbol interference and channel. As the key part of OFDM channel estimation technology has been the focus of attention, channel the estimation of the channel state information recovery becomes more accurate, the system performance has been improved. This paper focuses on the estimation method of OFDM in-depth research, for the classical estimation method deficiencies and put forward corresponding improvement algorithm, the research content and innovation are as follows in this paper. (1) studied the basic characteristics of wireless channel fading, analysis in the actual environment of wireless channel. The signal received through inter symbol interference and channel in wireless channel interference. Status of the principle of OFDM and its key technology, which introduces three kinds of structure of OFDM pilot and three kinds of estimation algorithms. Based on how to obtain the good channel response values of several research practice common interpolation algorithm. (2) studied the estimation method based on improved DFT threshold. After the signal after discretization often caused by spectrum leakage, the channel appeared in the proposed adding Hamming window to solve the problem. For the classical DFT estimation algorithm can not eliminate the problem of noise within the cyclic prefix, puts forward some improved threshold denoising method. The threshold method can effectively eliminate channel burst large pulse noise, and MATLAB simulation experiment of the algorithm. The experimental results show that the improved algorithm in the bit error rate and the minimum mean square error is the most Small, so the improved algorithm can effectively enhance the estimation performance. (3) studied the improved DFT estimation method in wavelet based on the traditional DFT-MMSE estimation algorithm is proposed. With improved wavelet less than DFT and threshold fusion algorithm shortcomings also puts forward improved algorithm using wavelet denoising and threshold denoising the fusion of wavelet decomposition. The algorithm is added by using the soft threshold method to remove noise signal after LS channel estimation, followed by inverse discrete Fourier transform again after threshold processing to a deeper degree of reduce noise on the signal. The simulation results show that the two algorithms can be very good to reduce noise, improve the performance of the system.

【学位授予单位】：江西理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN929.53

【参考文献】