Research On Trajectory Privacy Protection In Location Based Service

In recent years, with the rapid development of mobile Internet and the positioning equipment,LBS (Location Based Service) has become a part of people’ life. With the enhancement of privacyprotection concept, people are concerned about whether their location privacy is revealed when theyenjoy the service. Trajectory privacy is an aspect of the location privacy. The trajectoryK-anonymity and constructing none-service area on the motion path are mostly used in the existingtrajectory privacy protection methods. In this thesis, trajectory privacy protection technologiesbased on K-anonymity are further researched.Firstly, this thesis proposes a greedy based location K-anonymity algorithm. The algorithmtakes the distribution of the target users and their neighboring users into account; chooses optimalneighboring region in every step when constructing anonymous region, removes redundancy of theanonymity area which were constructed by greedy method, and then forms the final anonymousregions. Through experiments, this thesis compares and analyzes the stability and the accuracy ofthe algorithm, also analyzes the impact of the grid accuracy on the algorithm.Based on pseudonym mechanism, footprint mechanism and the greedy based locationK-anonymity algorithm presented in this thesis, this thesis proposes a privacy degree requirementadapted trajectory privacy protection scheme. This scheme can not only protect the user’s trajectoryprivacy, but also meet the different needs of privacy degree. The scheme divides the applicationspace into sensitive areas and non-sensitive areas. When user is moving in the non-sensitive areas,the anonymous server processes the user’s identity information by pseudonym, and when user ismoving in the sensitive areas, the anonymous server not only processes the user’s identityinformation by pseudonym, but also processes user’s location information by location K-anonymity.In the process of constructing K-anonymity, the scheme uses footprints to increase the difficulty ofthe attack. The impacts of pseudonym’s lifetime and the number of sensitive areas on the trajectoryprivacy protection degree are analyzed through theory and experiment.Theoretical analysis and experimental results show that the algorithm and the scheme presentedin this thesis have good effects and higher efficiency for trajectory privacy protection in locationbased services.