An Enhanced Method of Trajectory Privacy Preservation Through Trajectory Reconstruction

Abstract

Trajectory data of mobile users contain plenty of sensitive spatial and temporal information, and can support many applications through data analysing and mining. However, re-identification attack and inference attack on such data may cause serious personal privacy leakage. Existing privacy preserving techniques cannot protect trajectory privacy well or largely scarify data utility. In view of these issues, in this paper we propose an enhanced trajectory privacy preserving method which can protect the trajectory privacy preferably while maintaining a high utility of the trajectory in data publishing. A mechanism is proposed to protect the privacy through replacing stop points in the trajectory and an effective trajectory reconstruction algorithm is introduced to avoid the mutations of trajectory, and also deal with the possible presence of obstacles around trajectories. The performance of our proposal is comprehensively evaluated on a real trajectory dataset. The results show that our method achieves a high privacy level and improves the utility of trajectory data greatly, compared with the state-of-the-art method.