New Blind Evoked Potentials Separation Algorithm Based On Minimum Dispersion Criterion

Alpha stable distribution processes can model the impulsive random signals and noises well in physical observation. Traditional evoked potentials (EP) analysis is developed under the condition that the background noises in EP are Gaussian distributed. In this paper, we propose neural network structures related to multilayer feed-forward networks for performing blind EP separation based on the fractional lower order statistics (FLOS) and minimum dispersion (MD) criterion. The simulation experiments show that the proposed new algorithm is more robust than the conventional algorithm.