Underdetermined blind source separation using modified subspace method based on convex model

Abstract

This paper provides a convex model based subspace projection method with enhanced functionality, which can be used for Underdetermined Blind Source Separation (UBSS). The model takes into account both projection and size of the signal's subspace, without estimating the source number at Time-Frequency (TF) point. Simulation results show that it overcomes the shortage of conventional subspace method and achieves high separation performance. The proposed method can be employed as a preprocessing technology for audio separation and enhancement, as well as biomedical images, etc.