A Personal Classification Method Using Spatial Information of Multi-channel EEG

Biometric authentication using various biological information is studied by many researchers. We study a feature extraction method available for personal authentication by focusing on EEG in the biological information. In addition, since electroencephalograph technology has advanced significantly in recent years, multichannel EEG is possible to be relatively easily measured. Therefore, in this paper, as EEG features, we propose a method using a cross-correlation between electrodes obtained from the multi-channel electroencephalograph. In validations, a feature combination, which is obtained from the proposed method and time-frequency analysis, is used. A personal classification is performed by applying SVM to obtained features. Moreover, by detailed validations about the proposed method, we evaluate the possibility of the crosscorrelation between electrodes as features for the personal authentication.