Itus: an implicit authentication framework for android

Abstract

Security and usability issues with pass-locks on mobile devices have prompted researchers to develop implicit authentication (IA) schemes, which continuously and transparently authenticate users using behavioural biometrics. Contemporary IA schemes proposed by the research community are challenging to deploy, and there is a need for a framework that supports: different behavioural classifiers, given that different apps have different requirements; app developers using IA without becoming domain experts; and real-time classification on resource-constrained mobile devices. We present Itus, an IA framework for Android that allows the research community to improve IA schemes incrementally, while allowing app developers to adopt these improvements at their own pace. We describe the Itus framework and how it provides: ease of use: Itus allows app developers to use IA by changing as few as two lines of their existing code - on the other hand, Itus provides an oracle capable of making advanced recommendations should developers wish to fine-tune the classifiers; flexibility: developers can deploy Itus in an application-specific manner, adapting to their unique needs; extensibility: researchers can contribute new behavioural features and classifiers without worrying about deployment particulars; low performance overhead: Itus operates with minimal performance overhead, allowing app developers to deploy it without compromising end-user experience. These goals are accomplished with an API allowing individual stakeholders to incrementally improve Itus without re-engineering new systems. We implement Itus in two demo apps and measure its performance impact. To our knowledge, Itus is the first open-source extensible IA framework for Android that can be deployed off-the-shelf.