Research On Trajectory Privacy Protection Technology For Data Publishing

With the rapid development of positioning technology and its wide application, a lot of trajectorys data has been produced. Analyzing and mining the trajectory data can support application associated with moving objects, such as planning transportation networks. However, pulishing original trajectorys will treat individuals’ privacy. Therefore, in the data publishing how to satisfy the requirements of the users’ privacy requirements, and how to make the data publishing has higher availability become the hot studying field now.Now in order to solve this problem, some researchers have started working on it and achieved some results. Among these studies trajectory k- anonymization is a good method to solve this problem. But the traditional trajectory k- anonymization did not consider the road network constraints, so leading to poor trajectory anonymous effect, published trajectory has a data migration form road network and the data availability is low.To solve the above problems, this paper proposed a method that is trajectory k-anonymization privacy protection method based on road network. In this paper, the main work is as follows:(1) On the basis of the existed research, this paper propose a trajectory k-anonymization privacy protection method based on road network. This method improved the existed trajectorie equivalence partitioning method by putting the similar time and similar direction of trajectories into a same equivalent class set. Then considering the constraints of trajectory from the road network, this paper propose a measurement method of trajectory similarity. According to this measurement method, we construct space-time similar trajectory k-anonymization set of the trajectory in the equivalent class by the improved DBSCAN clustering algorithm. Finally we anonymous the trajectories within k-anonymization set by disturbing algorithm based on publishing atomic trajectory. The trajectorys in k-anonymization set don’t be distinguished between each other, so we can achieve the goal of trajectory privacy protection.(2) We designed and developed the trajectory data publishing system based on ArcGIS with the proposed algorithm in this paper. There are some functions in this system. Such as network data processing, trajectories anonymous, data usability evaluation, show and query original trajectories data and anonymous data etc.(3) In order to verify the effectiveness of the method, first proving the privacy protection ability of this method, the analysis shows that the proposed method is greatly improved the ability of privacy protectionthis, then proposed a usability evaluation method of trajectory data. We call it traffic flow distortion. Then we make a comparison between the the classical trajectory k-anonymization with the existing usability evaluation method and the proposed measurement method. The experimental results show that trajectory k-anonymization privacy protection method based on road network is better than the classical algorithm in the data usability.