Identifying Individuality Using Mental Task Based Brain Computer Interface

摘要: In recent years, numerous Brain Computer Interface (BCI) technologies have been developed to assist the disabled. this paper, mental task based BCI is proposed for a different purpose: identify individuality of person. The idea on classification electroencephalogram (EEG) signals recorded when user thinks either one or two tasks. As individuals thought processes, would be appropriate individual identification. To increase inter-subject differences, EEG data from six electrodes are used instead one. Sixth order autoregressive features computed and classified by Linear Discriminant classifier using modified 10 fold cross validation procedure, which gave an average error 0.95% tested 400 patterns four subjects. Though method undergo further development obtain repeatable good accuracy; initial study has shown huge potential over existing biometric identification systems as it impossible faked.