Release of Trajectory Data based on Space Segmentation using Differential Privacy
Yongxin Zhao, Wanqing Wu, Chaofan Di · 2021 IEEE 20th International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom) · 2021
The applications of location-based services(LBS) are increasingly important in people's daily life. However, these services may lead to privacy leakage, which causes more and more users to worry about privacy issues. Although the differential privacy technology has solved some location privacy problems, most of the current models are still unable to resist complex background-knowledge attacks. In this paper, we propose a novel approach to protect privacy for trajectory based on prefix tree. Firstly, we use the Hilbert curve to split the trajectory position points which the optimal divisions are selected by the index mechanism, and obtain the center point of each area. Secondly, we build a noisy prefix tree to store location points, it should be noted that the nodes of the tree store the polymerization location points. Finally, an arithmetic distribution method is applied to the privacy budget, meanwhile the amount of noise is limited by a threshold. The proposed algorithm is compared with related algorithm on the original dataset. Experimental results show that the algorithm proposed in this paper improves data availability while ensuring data privacy.