Sources Detection Based on Clustering in Spatially Correlated Noise Fields
Bao Zhiqiang · Ceshi jishu xuebao · 2006
To deal with the spatially correlated noise,a detection method based on Fuzzy-c-Means(FCM) clustering algorithm is proposed,which uses canonical correlation coefficients of the joint covariance matrix as the feature to be classified.Three crucial issues are discussed for the application of Fuzzy-c-Means method to the source-number detection.Compared with the classical methods,the presented algorithm has better performance and angular resolution.The simulation results demonstrate the effectiveness and robustness of the proposed scheme.