Title :
Neural blind separation of complex sources by extended Hebbian learning (EGHA)
Author :
Fiori, Simone ; Piazza, Francesco
Author_Institution :
Dipt. di Elettronica e Autom., Ancona Univ., Italy
Abstract :
The aim of this paper is to present a nonlinear extension to Sanger´s generalized Hebbian learning rule for complex-valued data neural processing. A possible choice of the involved nonlinearity is discussed recalling the Sudjianto-Hassoun interpretation of the nonlinear Hebbian learning. Extension of this interpretation to the complex case leads to a nonlinearity called Rayleigh function, which allows for separation of mixed independent complex-valued source signals
Keywords :
Hebbian learning; adaptive signal detection; neural nets; principal component analysis; signal sources; Rayleigh function; Sanger´s generalized Hebbian learning rule; Sudjianto-Hassoun interpretation; complex sources; complex-valued data neural processing; extended Hebbian learning; mixed independent complex-valued source signals; neural blind separation; nonlinear extension; Blind source separation; Equations; Hebbian theory; Lagrangian functions; Neural networks; Neurons; Principal component analysis; Vectors;
Conference_Titel :
Circuits and Systems, 1999. ISCAS '99. Proceedings of the 1999 IEEE International Symposium on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-5471-0
DOI :
10.1109/ISCAS.1999.777578
