Underdetermined mixing matrix estimation by comprehensive application of cluster validity indexes  

作　　者：Wang Chuanchuan Jiang Lin Zeng Yonghu Wang Liandong 

机构地区：[1]State Key Laboratory of Complex Electromagnetic Environment Effects on Electronics and Information System,Luoyang 471003,China

出　　处：《The Journal of China Universities of Posts and Telecommunications》2020年第6期73-86,共14页中国邮电高校学报（英文版）

基　　金：This work was supported by the National Natural Science Foundation of China(61801480).

摘　　要：To solve the problem of mixing matrix estimation for underdetermined blind source separation(UBSS)when the number of sources is unknown,this paper proposed a novel mixing matrix estimation method based on average information entropy and cluster validity index(CVI).Firstly,the initial cluster center is selected by using fuzzy C-means(FCM)algorithm and the corresponding membership matrix is obtained,and then the number of clusters is obtained by using the joint decision of CVI and average information entropy index of membership matrix,then multiple cluster number estimation results can be obtained by using multiple CVIs.Then,according to the results of the number of multiple clusters estimation,the number of radiation sources is determined according to the principle of the subordination of the minority to the majority.The cluster center vectors obtained from the clustering operation of the estimated number of radiation sources are fused,that is the mixing matrix is estimated based on the degree of similarity of the cluster center vectors.When the source signal is not sufficiently sparse,the time-frequency single source detection processing can be combined with the proposed method to estimate the mixing matrix.The effectiveness of the proposed method is validated by experiments.

关 键 词：UBSS ESTIMATION of the number of radiation sources MIXING matrix FCM clustering CVI 

分 类 号：TN911.7[电子电信—通信与信息系统]