The Blind Source Separation Method Based on Wavelet Packet and Variational Bayes Independent Compoment Analysis

Combined with the respective advantages of wavelet packet analysis and variational Bayesian independent component analysis(VBICA),a blind source separation method based on wavelet packet and VBICA is proposed. The observation signals are decomposed by the wavelet packet and the obtained reconstruct wavelet packet coefficients constitute the new observation signals. The source signals are estimated by the VBICA method. The simulation results show that the proposed method is very effective. Even under noisy mixture at the very low signal-to-noise ratio,the satisfactory separation performance can be also obtained.