Abstract | ||
---|---|---|
This paper investigates new algorithms for blind source separation that use cumulants instead of nonlinearities matched to the probability distribution of the sources. It is demonstrated that separation is a saddle point of a cumulant-based entropy cost function. To determine this point we propose two quasi-Newton algorithms whose convergence is isotropic and does not depend on the sources distribution. Moreover, convergence properties remain the same when there is Gaussian noise in the mixture. |
Year | DOI | Venue |
---|---|---|
2000 | 10.1109/ICASSP.2000.861206 | ICASSP |
Keywords | Field | DocType |
convergence property,quasi-newton algorithm,saddle point,novel blind source separation,new algorithm,cumulant-based entropy cost function,blind source separation,gaussian noise,sources distribution,probability distribution,adaptive signal processing,cumulants,cumulant,convergence,entropy,cost function,newton method | Convergence (routing),Mathematical optimization,Saddle point,Computer science,Higher-order statistics,Algorithm,Cumulant,Probability distribution,Gaussian noise,Blind signal separation,Source separation | Conference |
ISSN | ISBN | Citations |
1520-6149 | 0-7803-6293-4 | 6 |
PageRank | References | Authors |
0.78 | 10 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Sergio Cruces | 1 | 206 | 19.05 |
L. Castedo | 2 | 65 | 10.88 |
A. Cichocki | 3 | 342 | 33.68 |