Blind Source Separation Using Virtual Sensors

This paper presents a new approach to enhance the quality of separated source in blind source separation problem. We consider the general case where the total number of sources is larger than the number of sensors and the number of sources is not known. The method is based on a principal point that we can use the signals of the ldquovirtual sensorsrdquo that do not exist in practice but generated synthetically. In the proposed algorithm, the number of actual and virtual sensors equals the number of sources. This algorithm deletes the musical noise effect generated by DESPRIT algorithm, so the quality of separated signals increases compared to DESPRIT.