基于单矢量水听器空间谱增强的MUSIC算法

针对单矢量水听器多重信号分类(Multiple Signal Classification,MUSIC)算法在低信噪比条件下,存在谱峰宽度变宽、估计精度变低等性能恶化的问题,提出一种基于单矢量水听器空间谱增强的改进MUSIC算法。该算法的基本原理是:单声源入射到单只矢量水听器上时,加一参考信号源,若两者方位相同,则同方位叠加使空间谱增强;若两者方位不相同,则两方位合成,使空间谱估计产生偏差。因此,改变参考信号源的方位,当参考信号源的方位与信号源的方位一致时,将使谱峰得到增强,此时空间谱达到最大值,其对应的角度即为信号源的方位角。仿真分析及实验数据处理结果表明,与常规MUSIC算法相比,该算法具有更尖锐的谱峰、更高的估计精度,能够实现更好的空间谱估计。

A modified MUSIC algorithm based on spatial spectrum enhancement of a single vector hydrophone

A modified Multiple Signal Classification (MUSIC) algorithm of single acoustic vector hydrophone is proposed by enhancing space spectrum under low SNR condition. The basic principle of the algorithm is described as follows:an auxiliary source is added when a single sound source is incident on a single vector hydrophone, if the azimuth of the auxiliary source is the same as the signal source azimuth, the space spectrum will be enhanced; if the two azimuths are not in the same, the composite azimuth of the two azimuths will make the spatial spectrum estimation exhibiting deviation. So when the auxiliary source azimuth is consistent with the signal source one, the spectrum peak will be enhanced and the spatial spectrum reaches the maximum, moreover this azimuth assured is just the signal source azimuth. The results of simulation and test data processing show that, compared with conventional MUSIC algorithms, this algorithm has sharper peaks and higher estimation accuracy, can better achieve spatial spectrum estimation.