TY  - JOUR
T1  - Performance Analysis of Adaptive Self-Normalized Radial Basis Function Neural Network  (ASN-RBF) For Blind Source Separation
AU - , D. Malathi AU - , N. Gunasekaran 
JO  - International Journal of System Signal Control and Engineering Application
VL  - 1
IS  - 3
SP  - 171
EP  - 179
PY  - 2008
DA  - 2001/08/19
SN  - 1997-5422
DO  - ijssceapp.2008.171.179
UR  - https://makhillpublications.co/view-article.php?doi=ijssceapp.2008.171.179
KW  - Adaptive self-normalized radial basis function neural network
KW  -blind source separation
KW  -gradient descent technique
KW  -temporal predictability
KW  -blind source separation
AB  - This study focuses on extracting the individual source signals from an artificially mixed signal. Number of signals involved is minimum 3 source signals. An adaptive self-normalized Radial Basis function network is developed for solving unknown source separation problems. The gradient descent optimization algorithm is applied to update the parameters in the generative model. The performance of the proposed network is compared with the model by using temporal predictability and it is illustrated with computer simulated experiments. The scaling problem in the Blind Source Separation using temporal predictability is eliminated by the proposed ASN-RBF network.
ER  -