Multiuser detection based on a robust approach of principal-factor method with determination of the number of sources

In this paper, we focus on the problems of array signal processing for multiuser detection. For multiuser detection, we propose a new approach of noisy blind source separation with determination of the number of sources. Our proposed approach is a practical technique to detect hidden sources from noisy array signals, which have wide and diverse range of amplitude and outliers. In order to accomplish the source detection, we developed a robust approach of principal-factor method, which is extended by the principal component analysis (PCA) and the standard principal-factor method (PFM). Moreover, a cross-validation technique is applied to determine the number of sources. The computer simulation results illustrate the performance and effectiveness of our proposed approach.