Research On Location Privacy Protection In Mobile Social Network Environments

In recent years,mobile Internet technology and positioning technology has been rapidly developed and widely used,mobile smart devices have been everywhere.Mobile users can enjoy location-based services through mobile smart devices anywhere and anytime,such as finding hotels that are closest to the current location.Based on large numbers of track data generated by mobile users,application providers and research institutions can produce better services for mobile users related applications through the analysis and mining of trajectory data,such as planning the transport network.However,if the original location of the user does not carry out privacy protection or direct publish the original trajectory data,may lead to personal sensitive privacy information leakage.Therefore,without influencing the service quality of user’s real-time location and the published service quality of user’s trajectory data,how to better protect the user’s privacy has become the focus of the current research.Based on the objective of improving the effectiveness of the algorithm and the ability to resist the attack,this paper studies the key technology of mobile user location privacy protection.Based on the trusted third-party centralized system architecture,the key technology of mobile user location privacy protection is studied.The research contents include:the effectiveness of dummy,the data encryption,the anti-attack ability of(k-δ)-anonymous and the analysis of differential privacy service quality and so on.The works of this paper is as follows:(1)Aiming at the shortcomings of the dummy of traditional randomized dummy generation algorithm,a dummy generation algorithm based on query context and motion model is proposed.The algorithm uses the number of interest points in the history log file and the road network topology as the query context and the motion model,and the dummy is more realistic and effective in time and space.And using the encryption algorithm based on Hilbert space transform encrypts the dummy set,and thus makes the user query data more secure.(2)Because the location privacy protection of the traditional(k-8)-anonymous trajectory data publishing is still unsatisfactory in front of strong background knowledge attackers,(k-Ψ)-anonymous trajectory data publishing privacy protection algorithm based on differential privacy is proposed.The algorithm constructs the uncertain threshold set Ψ with certain adaptive ability based on the network topology,which improves the indistinguishabilty of the k anonymous set,and carries out differential privacy perturbation of these anonymous trajectories by adding Laplacian noise.And the achieving a very good trajectory data publishing privacy protection effect.(3)In order to verify the validity of the two privacy protection algorithms,the security analysis of the two algorithms is carried out.The analysis shows that the algorithm proposed in this paper can better resist the specific attack of malicious attackers.Finally,the performance of the algorithm is analyzed based on the Oldenburg highway network.The results of experiment show that the two algorithms can improve the veracity of the dummy and the indistinguishability of the trajectory data.And under the condition that the data availability is not affected,it can protect the security of the location data.