Text Input Biometric System Design for Handheld Devices
Naif M. Alotaibi, Emmanuel Bruno, Michael Coakley, Alexander Gazarov, Vinnie Monaco, Stephen Winard, Filip Witkowski, Alecia Copeland, Peter Nebauer, Christopher Dirk Keene, Joshua Williams · 2014
The development of networks, especially the Internet, is changing the way we do computing. The addition of multiple devices, such as smartphones and tablets, gives us new ways to access information, but at the same time makes sensitive information more prone to be lost or used by unauthorized users. In an effort to respond to those new threats, keystroke biometrics is an area being extensively studied for continuously authenticating users on those devices. Keystroke biometrics is a promising solution that will help guarantee the security of sensitive information and device access. It uses the keyboard only and is invisible to the user. Keystroke biometrics takes advantage of the natural style and rhythm in which a user inputs characters on a soft or hard keyboard. Keystroke rhythms, area covered on a soft keyboard for a particular key, and timing can be measured to build a biometric pattern for identification and authentication. This paper describes user interaction with a touch screen device using a soft keyboard and the data that can be mined from a soft keyboard. We present the structure and technical details of our biometric keyboard for the Android platform and our data collection process in our test cases.