Abstract:

With the rapid development of communication network technology,the rapid improvement of intelligent electronic device functions and the rapid advancement of positioning technology,location-based services make people's daily lives more convenient.However,users' privacy information about their locations is facing threats that cannot be ignored.The existing continuous query-oriented location privacy protection methods ignore the semantic information contained in the user's movement trajectories,causing attackers to use that information to mine the user's behavior habits,personal preferences and other privacies.At the same time,the traditional fake trajectory privacy protection methods generate multiple fake trajectories to confuse the user's real trajectory,but the transfer of semantic location points in the fake trajectory does not conform to the user's behavior rules,so we propose a spatiotemporal correlation location privacy protection method with semantic information.This method combines the historical semantic trajectories of the user with the semantic information on the location to construct a user behavior model,and constructs false trajectories that conform to the rules of user behavior according to the transition probability and temporal-spatial correlation between semantic locations at adjacent moments in the model,thus achieving the purpose of confusing the real trajectory of the user.Finally,based on the real data set,the algorithm is compared with the existing algorithms,which shows that the algorithm can effectively reduce the risk of location privacy leakage in continuous query scenarios when the attacker has relevant background knowledge.

Key words: location based services, spatiotemporal correlation, privacy protection, semantic trajectory