Multi-Feature Authentication System Based on Event Evoked Electroencephalogram

Studies on electroencephalogram (EEG) signals have demonstrated that EEG signals of different subjects under the same experimental environment have identity differences. In order to better utilize EEG signals in identity authentication, a new analytical approach to study authentication system based on EEG signals is designed in this paper. To increase the stability of authentication signals, event evoked potentials are introduced in this study. Subjects' photos are used as stimulus source and subjects are divided into groups when signals are collected. To increase classification accuracy, several analytical methods are used in this study to collect features. A specific classifier is established for each subject to study the classification accuracy. The variation of classification accuracy under different thresholds is investigated. Moreover, the effect of subjects' genders on classification accuracy is also studied. Finally, the difference between the subjects' brain responses to their own photos and to others' photos is demonstrated in 2D brain maps.