A New Source Number Detection Algorithm Based on Fuzzy Clustering
基于模糊聚类的信源个数检测新算法

The research of source number detection is still open and challenge issue in array signal processing. The accurate estimation may be very essential to those high resolution direction finding algorithms. However, the traditional methods are sensitive to different noise fields and the performances of them are degraded in short snapshots. To overcome these drawbacks, an effective, accurate, robust detection method of the number of sources is proposed using the fuzzy-c-means clustering algorithm. The simulation results with two noise fields demonstrate the effectiveness and robustness of the proposed scheme.