Design and implementation of a RFID-based authentication system by using keystroke dynamics

Password authentication is so convenient that it is widely used for user identification. However, such password authentication method is not robust enough if the legal owner is unaware of it. In this document, we propose an integrated technique approach to enhance user identification. We adopt keystroke dynamics as a biometric to strength conventional password mechanism and keep these characteristic values into RFID cards as pattern template for user identification. Firstly, the pattern extraction process captures keystroke timing, and then calculates thresholds by statistical method. After completing pattern extraction, the system writes these characteristic templates into RFID cards for individuals. At last, we show the experiment results based on about 1600 sample records collected from 21 users of different typing ability. The results of FRR and FAR approached to 6.50% and 3.50% respectively. It shows that the password authentication system with keystroke dynamics could effectively improve system security.