天线阵列信号的盲分离研究

【作者】 弓喜英；

【导师】 吴援明；

【作者基本信息】 电子科技大学 ， 信号与信息处理， 2007， 硕士

【摘要】 盲源分离问题是从观测数据中分离出未知但相互统计独立信号的过程,混合过程假定为线性时不变系统。盲源分离是当前信号处理领域的一个研究热点,盲信号处理技术由于不要求已知源信号以及信道的先验知识,在数字通信、阵列信号处理、水声信号处理、生物医学信号处理、图像处理以及语音信号处理等方面都有重要的理论研究价值和广泛的应用前景。本论文主要针对天线阵列信号的特性,提出一种实现窄带阵列信号盲分离方法,该方法是在EASI算法的基础上进行改进,将窄带阵列信号成功的分离开来。本篇论文主要研究内容如下:(1)搜集、整理、总结了国内外盲信号分离的最新研究成果,介绍盲分离基本概念和理论,并介绍了几种常用的盲分离算法,为了全面衡量盲信号分离效果,文章引入几个新的性能评价指标。(2)分析天线阵列信号的数学模型,针对其的特性,提出了一种实现窄带阵列信号的盲分离方法,该方法首先将天线阵列接收的窄带时延混合信号变换为复数域的瞬时混合信号,再用复数域的盲信号分离算法对变换后的混合信号进行盲分离。通过计算机仿真证实了该方法的可行性。(3)在实际工程应用中,实际环境比较复杂,所以必须考虑实际环境对信号分离效果产生的影响,这样可以在实际工程中有指导地应用盲信号处理技术。通过对阵元个数、信噪比、半径/波长、阵列流形等仿真条件进行实验,找出了它们的变化规律。(4)盲信号分离技术的目的是为了提取信号,而波束形成技术也是为了这一目的。通过盲信号分离技术与波束形成技术进行仿真比较,从而发现盲信号分离技术的优势所在。传统的波束形成技术依赖于天线阵列的方向矩阵,而在实际工程中,无法的得到精确的天线阵列的方向矩阵,而盲信号处理技术并不需要这些,只需要接收信号即可分离。本文主要是针对窄带阵列信号进行盲分离的研究,对现有盲分离算法的改进,成功的分离了窄带阵列信号,并对实际环境中的阵列信号,进行了性能分析。更多还原

【Abstract】 Blind Source Separation (BSS) problem is to separate individual source signals from a set of mixed signals without the prior information about source signals and the channel, assuming the mixed process is linear and time-fixed. BSS has been currently an active-research area belonging to signal processing. Because blind signal processing techniques do not need prior information of the original source signals and channel, they have significant value in theory study and broad foreground in practical applications in the fields of digital communication, array signal processing, underwater acoustic signal processing, biomedical signal processing, image signal processing, speech signal processing, etc.This paper mainly researches on blind separation problem of array signal, and proposes a blind separation method for array signal. This method is based on EASI algorithm. The main contributions are as follows:(1) The recent research findings are collected、sorted and generalized, the basic theory and algorithms are introduced. In order to evaluate blind separation performance more completely, several performance indexes of blind signal separation are presented.(2) The received signal model for narrow band array signals is analyzed. According to feature of received signals, this paper proposes a blind separation method for narrow-band time-delay mixture signals. This method transforms the array narrow-band time-delay mixture signals received to complex-valued instantaneous mixture signals which can be separated by the BSS algorithm .Simulation results show that the method can be used to separate the received array mixture signals effectively.(3) In project application, real environment is very complex, so we must consider influence of real environment on signal separation performance. This can guide us to apply blind separation techniques in practice project. Through simulation and experiment to factors of real environment, we can find theirs rule and reason of algorithm’s failure.(4) The aim of BSS is to extract signals from mixed signals, so as beamforming technology. Through comparison BSS with beamforming technology, advantage of BSS can be found. Beam forming must depend on direction matrix of antenna array. But in practice precision direction matrix isn’t obtained. Blind signal separation techniques only need received signals to separate.This paper studies on blind separation of narrow-band array signals, and develops existing algorithm, successfully separates array signals and analyses influence of real environment on signal separation performance.更多还原