Electrical and Computer Engineering
2332 Main Mall
Vancouver,
BC
Canada
V6T 1Z4
Tel +1.604.822.2872
Fax +1.604.822.5949
Email:
info@ece.ubc.ca
| Title | A New Diagonal Hessian Algorithm for Blind Signal Separation |
| Publication Type | Conference Paper |
| Year of Publication | 2006 |
| Authors | Elsabrouty, M., T. Aboulnasr, and M. Bouchard |
| Conference Name | Acoustics, Speech and Signal Processing, 2006. ICASSP 2006 Proceedings. 2006 IEEE International Conference on |
| Pagination | V -V |
| Date Published | may. |
| Keywords | blind signal separation, blind source separation, diagonal Hessian algorithm, Hessian matrices, maximum likelihood estimation, maximum-likelihood blind separation cost function, Riemannian geometry, second order optimization, speech processing, speech signals |
| Abstract | A new algorithm for blind signal separation of speech signals that does not require pre-whitening is proposed in this paper. The algorithm is based on second order optimization using Riemannian geometry. The algorithm employs several practical approximations to the Hessian matrix of the maximum-likelihood blind separation cost function, to produce a computationally efficient algorithm that is capable of working on-line. Simulation results show the improved performance of the proposed algorithm with different mixing data |
| URL | http://dx.doi.org/10.1109/ICASSP.2006.1661363 |
| DOI | 10.1109/ICASSP.2006.1661363 |