Research On Real-time Trajectory Privacy Protection Method Based On Semantic Information

With the popularity of location-based services in mobile web applications,users are facing the risk of privacy disclosure while enjoying convenience.The large amount of data generated by mobile users when using location-based services has distinct personal characteristics,and it is easy for attackers to dig out the user's home address,occupation and other personal privacy from the location-based service data.At present,the research of trajectory privacy protection is still a hot topic in the field of privacy protection and has great research value.The existing trajectory privacy protection methods are mainly divided into two aspects: one is the privacy protection when publishing the collected trajectory data;the other is the real-time trajectory privacy protection of mobile users.However,the existing methods seldom consider the semantic information of location,the lack of quantification of location privacy,the redundancy of anonymous area constructed by anonymous method,and so on.Therefore,the above problems,the author of this paper proposes a combining false position and K-anonymous tracks of privacy protection model,the main research contents include: various models of the first research trajectory of privacy protection,real-time trajectory privacy protection for further study,considering the real road network circumstances real-time trajectory privacy key points,ask questions and formalization;Secondly,mobile users are likely to disclose their location when using location-based services,and the existing methods lack the semantic information of location,so the privacy protection is weak.Therefore,a pseudo-location generation algorithm based on semantic information that can be completed in mobile users is proposed.The algorithm considers the frequency of mobile users' access to different types of POI,and USES semantic degree as the evaluation criterion to select false location points to confuse the real location of mobile users.Then delve into the track of privacy protection method is commonly used in K-anonymous model,in view of existing problems anonymous area redundant anonymous methods,put forward based on the K-anonymous users nearby algorithm(KAAU)and K-anonymous algorithm based on historical data(KAHD),the anonymous model considering the distribution of the semantic information of position and adjacent to the user density,advantage of the characteristics of K d tree data structure to find the location of the mobileuser is close,and the distance entropy and the degree of semantic as constraint condition,structure size and area of the anonymous.In this paper,the trajectory data generated by the simulator is used as experimental data to design experiments to analyze the effectiveness of the pseudo-position generation algorithm,and the k-anonymous algorithm is compared with the existing similar methods to verify the effectiveness of the algorithm by using the distance entropy and the size of the generated anonymous region.Finally,experiments show that the pseudo-location generation algorithm has a high degree of privacy protection,and the anonymous region generated by KAAU algorithm and KAHD algorithm is also significantly reduced.