Trajectory Privacy Preserving Methods For Moving Objects

At present,various kinds of mobile positioning devices are common in people's daily life,but the trajectory data generated during use is also secretly collected by third-part servers.These trajectory data contain rich personal information,However,if the data is directly released without any processing,it will leak the users' personal privacy.At present,hot topics of moving objects trajectory privacy protection include data availability,privacy model,and user personalization.Through the research and analysis of hot topics,this thesis proposes different trajectory privacy protection methods for these three aspects:(1)Based on the high availability of trajectory data,this thesis proposes a path-based privacy protection method called TOPF based on frequent paths.Firstly,a new set of comparison rules for frequent paths is defined.Then,under the condition of following the constraints of the road network,the k-anonymous trajectory groups are constructed by using frequent trajectory.Finally,the representative trajectory is selected from the first f frequent trajectories in the trajectory group.The trajectory with the highest similarity is released as a representative trajectory of the reorganization.The experimental results show that the TOPF method not only effectively guarantees the privacy of the user,but also ensures the availability of data.(2)Aiming at the problem that the current privacy model relies on the background knowledge of the attacker,.In this thesis,we address this problem and propose a Sequence R(SR)-tree structure which is based on the R-tree and under differential privacy.At first,we construct the SR-tree by using the trajectory sequence instead of the minimum bounding rectangle of the R-tree.Then we propose an attack model called non-location sensitive information attack,in order to resist this attack,we use differential privacy model to add noise into the location data and non-location data.Finally,since the data inconsistency may happen after the noise is added,so we use method to solve this problem and increase the utility of the data.Experimental results show that our algorithm not only has high data utility,operational efficiency,but also has good scalability.(3)In view of the inconsistency of users' privacy requirements,an improved trajectory privacy protection model is proposed for the users' personalized problem.The model has five parameters,including short-term disclosure,long-term disclosure,trajectories distance deviation,trajectories local similarity and services request probability,users can adjust these parameters to meet their own requirements through their own needs;Secondly,this thesis also designed a dummy trajectory generation algorithm for this privacy model to ensure the privacy of users.The experimental results show that the method of this paper not only effectively protects the user's personal privacy,but also satisfies the individual requirements.