Title :
A novel method to estimate of mixing matrix under over-complete cases in wavelet packet domain
Author :
Mozaffari, B. ; Tinati, M.A.
Author_Institution :
Fac. of Electr. & Comput. Eng., Univ. of Tabriz, Tabriz
Abstract :
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.
Keywords :
blind source separation; wavelet transforms; blind source separation; linear spectral bisecting; mixing matrix; over-complete cases; scatter plots; variance calculations; wavelet packet domain; Additive noise; Blind source separation; Independent component analysis; Laplace equations; Matrix decomposition; Source separation; Sparse matrices; Wavelet domain; Wavelet packets; Wavelet transforms;
Conference_Titel :
Computer and Communication Engineering, 2008. ICCCE 2008. International Conference on
Conference_Location :
Kuala Lumpur
Print_ISBN :
978-1-4244-1691-2
Electronic_ISBN :
978-1-4244-1692-9
DOI :
10.1109/ICCCE.2008.4580745
