基于独立分量分析的波束空间波达方向估计算法  

Beam-space Direction of Arrival(DOA) algorithm based on independent component analysis

作　　者：陈晋央[1,2] 吴瑛[1] 

机构地区：[1]解放军信息工程大学信息工程学院,郑州450002 [2]总参工程兵科研三所,河南洛阳471023

出　　处：《声学技术》2012年第2期227-231,共5页Technical Acoustics

摘　　要：波束空间变换是波达方向估计算法中重要的预处理方法,该方法适合大阵列、小信号源数的场合,可以有效降低运算量。但是目前存在的波束空间类DOA估计算法大多需要预知信号的大致来向,同时对阵型有要求。为了解决上述问题,提出一种基于独立分量分析的波束空间DOA估计算法,使用独立分量分析方法获得波束形成变换矩阵。该算法对阵型无要求、无需知道信号来向并且具备较强的角度分辨能力。仿真实验验证了有效性。Beam-space transformation is an important preprocessing of direction of arrival （DOA）, which fits for the large array and small source number situation and is able to reduce the computational complexity, but the existing beam-space DOA algorithms need the approximate directions of signals and cannot fit for all array. In this paper a beam-space DOA algorithm based on independent component analysis ICA） is proposed to solve the above problems, the algorithm gets the beam-space transformation matrix using ICA and has high resolution, which has no request for array manifold and approximate directions of signals. Simulation result reveals that the proposed algorithm has good performance.

关 键 词：波达方向估计 波束空间 独立分量分析 均匀圆阵 

分 类 号：TN911.7[电子电信—通信与信息系统]