Model for adaptable context-based biometric authentication for mobile devices

Abstract

It becomes possible to take advantage of seamless biometric authentication on mobile devices due to increasing quality and quantity of built-in sensors, increasing processing power of the devices, and wireless connectivity. However, practical effectiveness of the biometric authentication application depends on user's environment conditions that can decrease the accuracy of biometrics recognition or make the acquisition process undesirable for mobile user in a given moment, i.e., effectiveness depends on usage context. In this paper, context-based biometric authentication model for mobile devices is proposed. It enables determining the most accurate authentication method at the moment along with the most accurate form of interacting with a user w.r.t. authentication process. The generic model designed and verified with proof-of-concept implementation constitutes a foundation for building further adaptable and extensible multi-factor context-dependent systems for mobile authentication.

Don't Use Fingerprint, it's Raining!

“…Google's contextaware authentication on Android, Smart Lock [8], allows users to choose contexts in which their phones stay unlocked. Wójtowicz and Joachmiak [32] presented a generic model that allows choosing the "optimal biometrics" for mobile authentication based on contextual factors (e.g., no voice biometrics in silent mode).…”

“…However, switches to fallback are tedious. We further learn that context can be considered for authentication (e.g., location, proximity or other sensor values [11,14,17,30,32]) and that it is technically possible (e.g., choosing biometric mechanisms based on context [32]). Thus, several novel authentication mechanisms evolve (cf.…”

“…Factors include environmental properties (e.g., location), but also human factors [28]. Context can be leveraged to a) skip authentication in certain situations (e.g., Google Smart Lock [8]) or b) choose adequate (e.g., biometric) authentication [32].…”

“…While related work suggests that this is technically possible [32], we look into how context-aware selection of authentication is used and perceived by mobile users in the wild. In a two-week field study (N=29), we tested an Android application, suggesting users of fingerprint authentication to switch to their knowledge-based fallback based on (partly fake) context information.…”

Technical feasibility of context-aware passive payment authorization for physical points of sale

“…Such values are calculated by Trans component. The criteria related to the situation context influencing the usefulness and quality of AMs have been used in the model for adaptable context-based biometric authentication dedicated for mobile devices and described in one of our previous articles [17].…”