阵列方向矩阵Moore-Penrose逆的波达方向估计  

Direction-of-Arrival Estimation of Arry Steering Matrix Moore-Penrose Inverse

作　　者：黄紧德[1,2] 孙洪[2] 

机构地区：[1]广西教育学院数学与信息科学学院,广西南宁530023 [2]武汉大学电子信息学院,湖北武汉430072

出　　处：《信号处理》2016年第7期842-848,共7页Journal of Signal Processing

基　　金：广西高校科学技术研究项目(YB2014587)

摘　　要：针对弱信号且源数未知情况下的波达方向估计问题,传统方法是先估计源数目再进行波达方向估计,但源数目估计的误差会造成波达方向估计精度下降。本文提出一种阵列方向矩阵的Moore-Penrose逆和MUSIC(MUltiple SIgnal Classification)算法的联合波达方向估计方法,充分利用Moore-Penrose逆的盲源数波达方向估计和MUSIC算法的高精度的优势。该方法无需预先估计源数目就可进行波达方向估计。通过仿真实验并与传统方法比较,表明本文提出的方法具有更高的精度,特别是在低信噪比情况下,具有更强的鲁棒性。To estimate the DOA of weak signals and unknown source number,the traditional methods were to find the direction of arrival after estimated the source number,which led to the performance deterioration caused by incorrect source number estimation. In this paper,a direction of arrival estimation algorithm combined with the array steering matrix MoorePenrose inverse and MUSIC（ MUltiple SIgnal Classification） was proposed,made full use of advantage of the Moore-Penrose inverse that could estimate the direction of arrival of unknown source number and MUSIC algorithm with high accuracy.This method did not need source number for direction finding,thus avoided to suffer performance degradtion because of incorrect source number. To verify the performance of the algorithm,we has a mass of simulation testify. The results demonstrate： compared with the traditional methods that need source number estimation before find the direction of arrival,the propose method has higher accuracy,stronger robustness especially in low signal-to-noise ratio.

关 键 词：波达方向估计 MOORE-PENROSE逆 独立分量分析 多重信号分类 

分 类 号：TN911[电子电信—通信与信息系统]