Research On Anonymity Models And Algorithms For Resisting The Attack Of Sub-trajectory

With the development of positioning and mobile communication technologies, it becomes more convenient to get location data. Moreover, with the progress of navigation. LBS (location-based service) and moving objects tracking, a large number of trajectory data come into being. These data play an important role in city planning, intelligent transportation and the analysis of behavior patter etc. However, publishing and sharing these data will threaten users’privacy. Consequently, the research on privacy preservation in data publishing or sharing is necessary and imperative. Nowadays, Anonymity-model, Anonymity technology and Anonymity algorithms are hot topics in this area. The Anonymity-model aims to anonymity constraints meeting specific privacy requirements, and the Anonymity technology is a method to implement anonymity models, While Anonymity algorithms are procedures to implement anonymity model by employing anonymity technology. This thesis mainly concentrates on the research of anonymity-models and anonymity-algorithms. The main contributions are as follows:(1) In allusion to the inefficiency of SEQANON algorithm in implementing km-anonymity, the thesis proposes GREANON and ACGREANON algorithms. From the views of data availability and running time, the thesis compares the proposed algorithms with SEQANON algorithm. Experiment shows that the running time of these three algorithms is similar. However, GREANON and ACGREANON algorithm can generate anonymous data whose availability is higher than this one generated by SEQANON algorithm.(2) In allusion to the disclosure risk of sensitive location which exists in existing anonymity models, the thesis proposes a (c, k) m-anonymity model, which can preserve identity and sensitive location effectively by setting different c constraint values for sensitive position. The thesis also proposes an anonymous algorithm to implement (c, k) m-anonymity. The thesis evaluates the effectiveness of the proposed algorithm and the proposed anonymity model from the view of information loss and running time. The experiment shows that (c, k) m-anonymity model can resist the attack of identity and sensitive location effectively.(3) In allusion to the problem of low data availability that exists in anonymous algorithms based on the greedy strategy, the thesis proposes an anonymous trajectory algorithm based on the Genetic Algorithm. In this thesis, the proposed algorithms are evaluated from the view of information loss and running time. By comparing with greedy algorithm, the experiment shows that the trajectory algorithm based on genetic algorithm can improve the data availability of anonymous data.