Protecting Movement Trajectories Through Fragmentation

Abstract

Location-based applications (LBAs) like geo-social networks, points of interest finders, and real-time traffic monitoring applications have entered people’s daily life. Advanced LBAs rely on location services (LSs) managing movement trajectories of multiple users in a scalable fashion. However, exposing trajectory information raises user privacy concerns, in particular if LSs are non-trusted. For instance, an attacker compromising an LS can use the retrieved user trajectory for stalking, mugging, or to trace user movement. To limit the misuse of trajectory data, we present a new approach for the secure management of trajectories on non-trusted servers. Instead of providing the complete trajectory of a user to a single LS, we split up the trajectory into a set of fragments and distribute the fragments among LSs of different providers. By distributing fragments, we avoid a single point of failure in case of compromised LSs, while different LBAs can still reconstruct the trajectory based on user-defined access rights.

In our evaluation, we show the effectiveness of our approach by using real world trajectories and realistic attackers using map knowledge and statistical information to predict and reconstruct the user’s movement.