Title: Blind dependent sources separation method using wavelet
Authors: Rui Li; Fasong Wang
School of Sciences, Henan University of Technology, Lianhua Street, 450001 Zhengzhou, China.
National Key Laboratory for Radar Signal Processing, Room 1510, Xidian University, 710071, Xi’an, China; China Electronics Technology Group Corporation, The 27th Research Institute, Zhengzhou, Henan, China
Abstract: The purpose of this paper is to develop novel Blind Source Separation (BSS) algorithm from linear mixtures of dependent sources signals. Most of the proposed algorithms for solving BSS problem rely on independence or at least non-correlation assumption of source signals. However, in practice, the latent sources are usually dependent to some extent. On the other hand, there is a large variety of applications that require considering sources that usually behave light or strong dependence. The proposed algorithm is developed based on the wavelet coefficient representations using Continuous Wavelet Transformation (CWT) which only requires slight differences in the CWT coefficient of the considered signals in the same scale. Moreover, the proposed algorithm can extract the desired signals in the overcomplete conditions. Simulation results show that the proposed algorithm is able to separate the dependent signals and yield ideal performance.
Keywords: BSS; blind source separation; statistically dependent; CWT; continuous wavelet transform; ICA; independent component analysis; overcomplete; dependent source signals.
Int. J. of Computer Applications in Technology, 2011 Vol.41, No.3/4, pp.296 - 302
Available online: 25 Sep 2011