Research On Privacy Protection Based On Generating Virtual Trajectories In Social Networks

With the development of science and technology and the popularization of smart devices,information services have also developed rapidly,and at the same time,serious privacy leaks have also been caused.Faced with the issue of privacy disclosure,research related to privacy protection has emerged.The existing trajectory privacy protection method is mainly a fake trajectory method.The hidden area in the fake trajectory is based on querying users,and most of the fake trajectories used in the generation of the fake trajectory are virtual non-existent users.There are many challenges in location point generation,anonymous area construction and trajectory similarity:(1)In trajectory privacy protection,in order to have enough users to send messages at the current location and meet the K-anonymity principle,a virtual location point needs to be generated,but the location and information characteristics of the virtual location point do not match the real user,so there is no confusion the purpose of the attacker.(2)Constructing an anonymous area is an important method of privacy protection,but the center of the anonymous area is the user requesting real information,or the geographical environment of the anonymous area is quite different.With some background knowledge,the requested information is easy to leak.(3)The protection efficiency of the trajectory is proportional to the similarity of the trajectory.The fake trajectory is different from the real trajectory in the direction and speed of travel.Even if multiple fake trajectories are constructed,the privacy protection will not achieve the desired effect.The research in this paper has conducted research and study on the virtual location points,anonymous areas and trajectory similarity in the above trajectory privacy protection.The main research work and innovative methods are as follows:(1)For the anonymous area with users as the core,this paper proposes the LTC algorithm to prevent the anonymous area center from being a real user.In terms of anonymous areas,the CAR algorithm in this paper is to determine the K-anonymous circular anonymous area by querying users and collaborating users.In terms of generating false trajectories,a new CFT algorithm is designed in this paper,which uses collaborative users to generate virtual regions to generate more effective false trajectories.In the experiments of analyzing the anonymity and overhead of false trajectories,the experimental results prove that the trajectory privacy protection method is more effective than the existing methods.(2)Aiming at the issue of dynamic user information leakage,a new trajectory privacy protection method TFR-res is proposed.This method mainly relies on finding similar users and proposes the method of using wave distance and angle to generate similar location points instead of sending by real users Message,and construct a hybrid trajectory through the proposed SMTA method.Since there is no information requesting user in this trajectory,the expected privacy protection result is achieved.By using real data to realize the TFR-res method,it is proved that the proposed method has an average high efficiency for the privacy protection of trajectory privacy.(3)In this paper,a new IR-TRK trajectory privacy protection method is proposed to address the problems that the virtual location point and the user's location and the real user trajectory are not similar to the fake trajectory in the fake trajectory.This method introduces the concept of IR heat map,divides the heat level by past historical data,finds the area blocks of the same environment as the real user to generate virtual location points,and then generates false trajectories similar to the real user through the trajectory similarity problem.The experiments in this paper prove that our method is relatively effective in protecting privacy.