Multichannel EEG-based biometric using improved RBF neural networks

摘要: Electroencephalogram (EEG) brainwaves have recently emerged as a promising biometric that can be used for individual identification. In this study, we present new visual stimuli-driven, non-volitional brain responses based methodological framework towards The mechanism provides an even more secure way in which the individuals are not aware of security credentials and thus manipulate their activities. Given intercorrelated structure functional areas, instead making identification decision relying on any single EEG channel, propose approach decision-level fusion multichannel signals, using Radial Basis Function (RBF) neural network its improved versions. Specifically, is determined according to patterns reflected from multiple channels over desired region. We evaluate performance our proposed methods four different stimuli independent channels. Experimental results show that, technique significantly improve accuracy, compared conventional channel solution. For RBF network, accuracy identifying 37 subjects could reach 70%, better than average about 55% achieved through networks, frequency-based 90%, while probability-based method 91%. Our study lays foundation future investigation accurate reliable brainwave-based biometrics.