Research On Personalized Trajectory Privacy Protection Method For Location Based Service

With the popularity of mobile devices and the development of social networks,Location Based Service(LBS)has been widely used in people's daily lives.While users enjoy the convenience of LBS,it will generate a large amount of trajectory data,which contains rich spatio-temporal information and sensitive data.If it is not protected,mining and analysis at will will lead to large-scale personal privacy leaks and even cause society.problem.Therefore,the primary goal of trajectory privacy protection is to protect the privacy of mobile users' trajectory data,which is also an important research content in the field of information security.Existing trajectory privacy protection methods generally have a uniform privacy protection scheme for user trajectories,which lacks personalization and accuracy,resulting in poor privacy protection effects.At the same time,the existing differential privacy trajectory protection methods generally lack noise sequences.Considering the correlation with the user's original trajectory sequence and release trajectory sequence,it is easy for an attacker to use filtering methods to filter,so that the privacy of the user's trajectory is leaked.Aiming at the above problems,this paper proposes a personalized privacy level definition algorithm,and proposes a(,)-differential privacy protection model based on the ?-differential privacy protection model,which is assigned according to the privacy levels of different locations.Different differential privacy budget parameters,so as to provide personalized and accurate privacy protection for the user's original trajectory,and improve the effectiveness of privacy protection.Then,the idea of noise candidate set is proposed,and the trajectory cross-correlation constraint is combined to ensure that the added noise sequence and the original The trajectory sequence and the final release trajectory sequence have spatio-temporal correlation,which effectively avoids denoising methods such as attackers' filtering methods,and achieves a good privacy protection effect and data availability.The experiments on the real data set prove the feasibility of the method proposed in this paper.The method in this paper is directly compared with other similar methods on the same data set.The experimental results show that the method proposed in this paper effectively improves The privacy protection effect and the data availability of the release track have better practical application value.