Evaluation of the Brain Wave as Biometrics in a Simulated Driving Environment Isao Nakanishi Koutaro Ozaki Shigang Li Graduate School of Engineering, Department of Engineering, Graduate School of Engineering, Tottori University Tottori University Tottori University 4-101 Koyama-Minami, Tottori, 4-101 Koyama-Minami, Tottori, 4-101 Koyama-Minami, Tottori, 680-8552 Japan 680-8552 Japan 680-8552 Japan Email:
[email protected] Abstract—In the case of user management, continuous or the while that the application is executed but the continuous on-demand biometric authentication is effective for achieving authentication brings a heavy load to the system. In fact, it was higher security with a light system load. However, it requires reported that the overhead for the continuous authentication us to present biometric data unconsciously. In this paper, we focus attention on the brain wave as unconscious biometrics. was 42 % in Ref. [6]. In particular, assuming driver authentication, we measure the Thus, we have proposed on-demand authentication in which brain waves of drivers when they are using a simplified driving users are authenticated on a regular schedule or a nonregular simulator. We evaluate verification performance using 23 subjects one on demand of authentication from the system [7]. The and obtain the EER of about 20 %. on-demand authentication makes the system load lighter. By the way, the fingerprint and the iris are not suitable I. INTRODUCTION for the continuous and the on-demand authentication because Biometric person authentication gains public attention. It is they ask users to present biometric data whenever the users well known that the fingerprint and the iris achieve higher are authenticated.