一种用于无线通信DOA估计的改进MUSIC算法研究

摘要

The MUSIC algorithm is a kind of spatial spectrum estimation algorithm,which needs to spend a long observa-tion period to estimate the CSDM when the spatial spectra of broadband signals are estimated,and is adverse to the localization of fast-moving targets. A MUSIC algorithm based on the steered covariance matrix(STCM)is proposed in this paper,in which characteristic decomposition on each band′s CSDM is performed firstly,and then the STCM of the noise space is derived by means of the noise subspace of every band to get the spatial spectrum estimation result of the whole broadband signals by using the noise space′s STCM. Simulation results show that this algorithm needs short observation time,suits for low SNR and has smaller observed variance while provides high resolution.%MUSIC算法是一种空间谱估计算法，在对宽带信号进行空间谱估计时，该算法需要较长的观测时间来估计协方差矩阵，不利于高速运动目标的定位。提出了基于驾驶协方差矩阵（STCM）的MUSIC算法，该算法首先对每个频带的CS-DM进行特征分解，然后利用各频带的噪声子空间求得噪声空间的STCM，进而利用噪声空间的STCM直接得到整个宽带信号的空间谱估计结果。仿真表明该算法在保证高分辨率的同时，需要较短观测时间，适用于较低信噪比、具有较小观测方差。