A novel method to estimate of mixing matrix under over-complete cases in wavelet packet domain

Wavelet packets decompose signals in to broader components using linear spectral bisecting. Mixing matrix is the key issue in the blind source separation (BSS) literature especially in under-determined cases. In this paper, we propose a novel method to obtain the best wavelet packet node to find sources directions in scatter plots using variance calculations and therefore estimating mixing matrix from noise free linear mixtures in over-complete cases is done.