基于二元矢量阵MUSIC算法的相干源分辨研究

Research on the resolution in coherent sources using a MUSIC algorithm based on a dual vector-sensor array

研究了单矢量传感器多重信号分类（SV-MUSIC）算法对非相干源的分辨性能，推导了声源等功率情况下的信噪比分辨门限表达式。针对相干源的分辨问题，提出了一种基于空间平滑的二元矢量阵多重信号分类（DV-MUSIC）算法。仿真实验表明，SV—MU-SIC算法对非相干源的分辨能力随着信噪比的增加而提高，在相应的判定准则下其结果与理论值基本吻合，而基于二元矢量传感器的DV—MUSIC算法能够实现相干源的分辨，其分辨性能曲线一定程度上类似于SV—MUSIC算法对非相干源的分辨性能曲线。

The performance of the single vector-sensor multiple-signal-classification （SV-MUSIC） algorithm for resolving incoherent sources was researched, and the expression of SNR resolution threshold when sound sources are equal in power was deduced. And then for coherent sources, a dual vector-sensor multiple-signal-classification （ DV-MU- SIC） algorithm was presented based on the spatial smoothing method. It was shown by simulation that, the resolu- tion performance of the SV-MUSIC algorithm for incoherent sources was improved as the SNR increasing, which meets the theoretical value under specified criterion. Based on a dual vector-sensor array, the DV-MUSIC algorithm was used to distinguish coherent sources and its resolution performance curve was similar to that of the SV-MUSIC algorithm for incoherent sources to a certain extent.