一种无特征分解的快速子空间DOA算法  

Fast subspace DOA algorithm without eigendecomposition

作　　者：彭巧乐 司锡才[1] 李利 

机构地区：[1]哈尔滨工程大学信息与通信工程学院,黑龙江哈尔滨150001

出　　处：《系统工程与电子技术》2010年第4期691-693,共3页Systems Engineering and Electronics

基　　金：国防基础科研基金(A242006110406)资助课题

摘　　要：基于子空间正交特性的MUSIC算法具有优良的超分辨性能,但由于其需要对空间协方差矩阵进行特征分解,因而计算量比较大。为了降低计算复杂度,提出一种快速子空间算法。该方法利用信号特征值大于噪声特征值的特性,通过对空间协方差矩阵的高阶次幂或者空间协方差矩阵逆的高阶次幂来逼近信号子空间或者噪声子空间,从而避免了特征分解。获得噪声子空间后再采用MUSIC算法实现波达方向估计。仿真结果表明,该方法减少了计算量同时能够达到MUSIC算法的估计性能。The MUSIC algorithm with sub-space orthogonal characteristics has an excellent super-resolution performance,but it needs to eigendecompose the spatial covariance matrix,which leads to a great computational cost.To reduce the computational complexity,a fast sub-space algorithm is proposed.Making use of the characteristic of signal eigenvalue being larger than noise eigenvalue,this method approximates the signal sub-space or noise sub-space through the high order power of the spatial covariance matrix or the inverse one to avoid the eigendecompsition.After obtaining the noise sub-space,it is capable to get the DOA by a MUSIC algorithm.The simulation result shows that the method achieves the performance of the MUSIC algorithm while reducing the computational cost.

关 键 词：到达角估计 超分辨 MUSIC算法 特征分解 

分 类 号：TN911.23[电子电信—通信与信息系统]