文档详情

基于压缩感知的稀疏多径信道估计综述.doc

发布:2017-04-19约1.12万字共9页下载文档
文本预览下载声明
基于压缩感知的稀疏多径信道估计综述 刘松1, ?作者简介:刘松(1986~),男(汉),河南睢县,研究生,研究方向为通信与信息系统。E-mail: songli 联系人摘要:压缩感知理论为信息处理领域带来了革命性的技术突破,引起了广泛关注。近年来,无线通信频谱资源日益紧张,将压缩感知理论与信道估计相结合以提高频谱资源利用率成为了新的研究热点。文章(本文)介绍了压缩感知的基本原理,综述了(分析了)基于压缩感知的单天线宽带稀疏多径信道估计的关键技术,着重阐述了其模型建立与信道响应的重构算法。与传统的信道估计方法相比,压缩感知的方法可用更少的训练序列而获得更佳的估计性能。文章(删除)最后对压缩感知理论应用于信道估计的最新研究进展作了介绍,评述了其中存在的某些问题,展望了未来发展方向。 关键词:压缩感知;稀疏多径信道;信道估计 A Survey On Compressed Sensing Based Estimation Of Sparse Multipath Channels LIU Song1,? (Information Engineering University . zhengzhou 450002) Abstract: The compressed sensing theory has brought the revolutionary technological breakthrough for the information processing domain, and arouses the widespread interest. Recently, spectrum of wireless communication becomes less and less, more and more people research for the channel estimation based on compressed sensing theory to increase operation rate of spectrum. This paper introduces the principles of compressed sensing, summarizes key technologies of compressed sensing based estimation of single-antenna wideband sparse multipath channels, elaborated its model building and the restructuring algorithm of channel response emphatically. Compared with the traditional channel estimation, the new method can get obtain the same estimation performance with less training sequence. Finally, the paper will review the development of compressed sensing theory and also consider its application in the field of wireless communication. Key words: compressed sensing; sparse multipath channel; channel estimation 近年以来,无线通信技术高速发展,人们对通信容量与可靠性的要求越来越高,于是频谱资源越来越显得稀缺。多径效应是无线通信的主要特点之一,通信系统中的相干接收需要利用均衡来准确校正多径传输信号,而可靠的均衡需要准确的信道状态信息。因此,信道估计是无线通信中较为关键的技术。 目前信道估计的方法主要有两类:盲信道估计方法和基于训练序列的信道估计方法。盲信道估计的方法由于其计算复杂度高,精度有限,不利于实际应用。目前无线通信系统中常用的信道估计方法是基于训练序列的估计方法。此方法通过发送训练序列,在接收端利用最小二乘(LS)或最小均方误差准则估计出训练序列处的信道响应值,再用插值算法来完成整个信道响应的估计。但传统的基于训练序列的估计方法并没有考虑信道本身的特性,若为获得较高的估计精度,必须插入更多数量的导频,这直接导致了频谱资源的极大浪费。 事实上,越来越多的物理验证与经验证据表明:许多无线信道的冲激响应存在稀疏特性结构,如高清数字电视(HDTV)信道、水声通信(UWA)信道以及超宽带(U
显示全部
相似文档