Continuous query in location-based services may reveal the attribute information of the user obliviously, and an adversary may utilize the attribute as background knowledge to correlate the real locations and to generate location trajectory. Thus, the adversary can obtain the personal privacy of the user. In order to cope with this problem, several algorithms had been proposed. However, these algorithms were mainly designed for snapshot query and failed to provide privacy protection service for continuous query. As a matter of fact, continuous anonymous regions can also be used as the trajectory of regions and one can obtain the real location trajectory through calibration. In addition, other algorithms designed for continuous query may also utilize a longer running time to achieve the attribute anonymity and affect the balance of quality of service and personal privacy. Therefore, in order to cope with the above two problems, this paper provides a PSO anonymization, short for particle swarm optimization anonymization algorithm. This algorithm utilizes the particle swarm optimization clustering algorithm to accelerate the process of finding similar attributes in attribute generalization. Furthermore, this algorithm also utilizes the randomly chosen anonymous cells to further generalize the anonymous region, so that it can provide better privacy protection and better service quality. At last, this paper utilizes security analysis and experimental verification to further verify the effectiveness and efficiency of both the level of privacy protection and algorithm execution.
Semantic and Trade-Off Aware Location Privacy Protection in Road Networks Via Improved Multi-Objective Particle Swarm Optimization
