摘要
提出一种宽带加权聚焦高分辨定向算法，用于解决宽带阵列信号各频率分量上信噪比不同时的高分辨定向问题．ＷａｎｇＨ提出的ＣＳＭ算法假设备频率分量上信噪比相同，而实际情况往往与此不符，因此在应用中性能不理想、本文在详细分析协方差矩阵结构的基础上，导出单源情况下利用协方差矩阵进行各个频率信噪比估计的表达式、然后进一步向多源情况推广．通过对各频率分量进行加权聚焦，最终改善方位估计性能．文中利用计算机仿真、水池实验及实测海洋舰船数据对该方法进行了详细验证，结论认为，加权聚焦算法能有效改善原方法的宽容性，估计精度和分辨率都有明显提高．
Wang H presented Coherent Signal-Subspace Method(CSM), which assumed that the signal-to-noise ratio(SNR) in all narrow-band components are equal. When this is not the easel its performance will degrade consequentlyIt is a natural approaCh to average the focused narrow-band covariance matrices proportional to the SNR at each frequency. In this paper, a robust weighted focusing matrices is proposed for CSM. An explicit expression is deduced toeStimate the SNR based on studying the characteristic of covariance matrices and a more reasonable focusing matrix isformed which can improve the performance of original CSM method in aPPlication. Computer simulation, water tankexperiment and ship noise data analysis results are shown to verify the efficiency of the method.
出处
《声学学报》
EI
CSCD
北大核心
2000年第1期16-20,共5页
Acta Acustica
