Quickest Intrusion Detection in Mobile Active User Authentication

In this paper, we address the problem of quickly detecting intrusions with lower false detection rates in mobile Active Authentication (AA) systems. Bayesian and Minimax versions of the Quickest Change Detection (QCD) algorithms are introduced to quickly detect intrusions in mobile AA systems. Furthermore, we introduce a new evaluation metric for comparing the performance of different AA systems. Effectiveness of the proposed framework is demonstrated using three publicly available unconstrained AA datasets. It is shown that the proposed QCD-based intrusion detection method can perform better than many traditional AA methods in terms of latency and low false detection rates.