Efficient Intrusion Detection for Mobile Devices Using Spatio-temporal Mobility Patterns

Sausan Yazji, Robert P. Dick, Peter Scheuermann, Goce Trajcevski

Abstract: Mobile phones are ubiquitous and are used for email, text messages, navigation, education, and as a pyment tool (e.g., Mobile Money - extensively used in China and Japan [1]). Consequently, mobile devices carry a lot of personal data and, if stolen, that data can be more important than the loss of the device. Most of the works on mobile devices security have focused on physical aspects and/or access control, which do not protect the private data on a stolen device that is in the post authentication state. However, some existing works, e.g. Laptop Cop [2] aim to protect data on stolen devices by remotely and manually deleting it, which requires user intervention. It may take hours before the user notices the loss of his device.