Biometrics Authentication Based on Spectrum Features of Electroencephalogram (EEG)

In this paper, we propose the biometric authentication using the electroencephalogram (EEG).The EEG is difficult to reproduce physically. Moreover, the authentication using the EEG makes possible to verify the user continuously. However, it is difficult to utilize the EEG as an individual feature. In this paper, in order to extract the individual feature, we analyzed the EEG in a frequency domain. As a result, it was found that the maximum value of a power spectrum and its frequency were different among users. Based on such a spectral feature, we obtained the verification rate of 85% through verification experiments in four users.