In this paper, a neural networks training based on Sequential Extended Kalman Filtering (SEKF) analysis for extraction and classifi cation of recorded EEG signal is proposed to improved feature extraction, classifi cation accuracy,and communication rate as well. The robustness of the SEKF against background noises has been evaluated by comparing the separation performance indices of the SEKF with well known algorithms (i.e., BPNN, JADE,and SOBI). A statistically signifi cant improvement was achieved with respect to the rates provided by raw data.
Copyrights © 2013