平面阵的方位估计算法及其误差分析

提出了一种基于信号相位匹配原理的平面阵方位估计的奇异值分解(SVDSPM)算法。推导了方位搜索需要的时延计算的一般公式,仿真分析了几种典型平面阵的方位估计性能,仿真结果表明:SVDSPM方位估计算法的性能优于MUSIC算法;理论分析了目标距离、传声器位置误差和接收通道幅度和相位不一致性等对定向性能的影响;以十字平面阵为例,仿真分析了上述几种因素对方位估计精度的影响,仿真结果表明:当传感器安装误差小于10 mm,接收通道幅度不一致性小于1.5 dB,接收通道相位不一致性小于10°和目标初始俯仰角大于45°时,方位估计的标准差小于0.5°。

A method of direction of arrival estimation using a plane array and analysis of its performance

A method of direction of arrival (DOA) estimation-SVDSPM (Singular Value Decomposition based on signal-phase-matching) algorithm with a plane array for long range targets is presented.The formula for calculating time delay that is necessary for estimating space spectrum is derived.Then DOA estimation performance for some typical plane arrays is analyzed and the simulation results preliminarily show that the performance of the method is more accurate than that of MUSIC (Multiple Signal Classification) algorithm at low ratio of signal to noise.Taking a cross plane array an example,the performance of the method under some influence factors,such as range,sensor position error and channel gain and phase uncertainties is theoretically analyzed.The simulation results show that when sensor position error,channel gain and phase uncertainties are smaller than 10 mm,1.5 dB and 10°respectively,and the angle of the pitch of target is smaller than 45°,the accuracy of DOA estimation is smaller than 0.5°.