单矢量传感器MUSIC算法的DOA估计及性能评价

针对单矢量传感器MUSIC算法在实际工程应用中存在性能恶化的问题,通过理论分析和数值仿真研究了幅相特性因素对其方位估计性能产生的影响,推导了幅相特性不一致条件下算法的空间谱估计表达式,同时给出了一种基于单矢量传感器通道功率盲归一化思想的改进算法.研究的结果表明:通道的幅相特性不一致将导致空间谱的谱峰变宽,并使空间谱估计的结果产生偏差甚至"伪峰";而改进的算法在保证方位估计精度的同时改善了目标的方位分辨能力,消除了传统算法空间谱中可能存在的"伪峰".湖试数据的处理结果进一步验证了改进算法的有效性.

Performance evaluation of DOA estimation using a single acoustic vector-sensor based on an improved MUSIC algorithm

The algorithm based on MUSIC using a single acoustic vector-sensor allowed for a new way to carry out DOA estimation,and the influence of the non-consistency characteristic of breadth or the phase on DOA estimation capability was deeply researched.The corresponding expression of spacial spectrum estimation was educed as well when the characteristics of breadth and phase are not consistent.In the meantime,an improved algorithm based on the idea of single acoustic vector tunnel power blind unification was given.The result of analysis in theory and simulation shows that the apex of the spacial spectrum was broadened in width and warp and even a ’fake peak’ appeared when the characteristics of breadth and phase are not consistent.The results of theory and simulation show that the resolution using the Improved MUSIC algorithm(IMUSIC) is better than the conventional MUSIC algorithm,it improves the direction distinguishing ability as well as keeping the DOA estimation precision.The ’fake peak’ is eliminated with a good precision at the same time,and the result of processing data during lake experiments testifies to the validity of the IMUSIC algorithm.