Skip to Main Content
Significant security concerns have come to the attention of both service providers and end users in cellular mobile systems. In these systems, misuse-based intrusion detection approaches are not effective since a potential wide variety of mobile users' behaviors are not taken into consideration. In this paper, by exploiting the location history traversed by mobile users, we propose an anomaly detection scheme to identify a group of especially harmful insider attackers - masqueraders. A realistic network model integrating geographic road-level granularities is proposed to effectively utilize users' location information. Based on this model, an Instance-Based Learning (IBL) technique is presented to construct mobile users' movement patterns. Simulation results demonstrate the effectiveness of the proposed scheme in terms of false positive rate and detection rate.