Privacy Risks in Publishing Mobile Device Trajectories

Growing availability of mobile devices capable of sensing and transmitting geospatial information has led to a wide range of location based services (LBSs). Sharing location data with others is an intrinsic feature of LBSs. However, publishing location may raise serious privacy concerns due to its close connection with users´ sensitive information. This work identifies a new privacy threat called trajectory_matching. We show that available privacy protection techniques preserving spatial relations of location samples are likely vulnerable to trajectory_matching in existence of background information. We develop a matching algorithm, trajectory_finder, and analyze its effectiveness on a real-life privacy protected trajectory dataset.