Secure Session on Mobile: An Exploration on Combining Biometric, TrustZone, and User Behavior

Abstract

With the rise of Internet connected mobile devices, applications have migrated from PCs to mobile computing platforms. An important aspect, payment processing, faces new security challenges from these developments. Inasmuch, these advancements demand efforts from researchers and industry to meet increasing security needs. Threats can ensue from data loss, theft from lost, stolen, or decommissioned devices, informationstealing malware, and password peeping. We propose a secure framework for sensitive session driven applications which combines biometric-based continuous and implicit tracking of user identities, and TrustZone. This framework is accomplished through monitoring fingerprint authentication logs as well as detecting events when the phone has left the user's hands, all while in TrustZone, a platform for secure computation and storage on mobile devices. This solution leverages multiple onboard sensors as well as the ARM architecture to accomplish these feats. We conducted two user-studies acquiring smartphone users' usage statistics to investigate security and usability needs of our identity-tracking solution. To monitor these subtle gestures in real-world uncontrolled environments, multi-session data collection has been conducted to iteratively improve system performance. The evaluation results have demonstrated the feasibility of this framework as a secure session-based payment system.