Research On Location Privacy Protection Method Based On Fake Location Under Continuous LBS Quer

With the rapid innovation of location technology and the upgrading of mobile intelligent devices,Location Based Services(LBS)have been widely used and have become an indispensable part of people’s daily lives.As the number of LBS users explodes,the data of users gradually contains significant commercial value,which makes personal data leakage a threat for users when using LBS.Therefore,how to protect the location privacy data of users has become an urgent problem in LBS.As a result,many location privacy protection mechanisms have emerged.Among the existing methods,the fake location-based method is currently the most commonly used LBS location privacy protection method as it does not require complex cryptographic techniques,does not rely on third parties,and can provide accurate query results to users.However,the existing fake location-based methods have shortcomings in continuous query scenarios.For example,as the number of user queries increases,the number of fake locations generated to effectively protect the user’s location privacy continues to decrease.Moreover,when generating fake locations,these methods do not consider user behavior attributes and spatiotemporal information between non-adjacent requests and positions during continuous user queries.In response to these issues,this article conducted research on fake location privacy protection methods for continuous LBS queries based on existing research.The research work and main results of this article include:(1)To address the problem of the decreasing number of fake locations that can effectively protect the user’s real location as the number of continuous queries increases,this article proposes a location privacy protection scheme based on query obfuscation for continuous LBS queries.The main idea is to confuse LBS providers by sending both real and fake queries,where the user submits a set of locations consisting entirely of fake positions,thereby disrupting the correlation between sets of locations submitted by the user for real adjacent queries,thus protecting the user’s location privacy.Through theoretical analysis and extensive experiments,this scheme not only efficiently generates location sets for users but also effectively prevents LBS providers from identifying fake locations by analyzing the correlation between sets of locations submitted during continuous queries.(2)Existing fake location-based methods did not consider user behavior attributes and spatiotemporal information between non-adjacent requests and positions during continuous user queries when generating fake positions.To solve this problem,this article proposes a fake position generation scheme based on spatiotemporal attention mechanisms.This scheme uses self-attention mechanisms to learn the relative spatiotemporal information between user continuous query trajectories and allows for interaction between non-adjacent requests and positions within the continuous query trajectories,making the generated fake positions and synthetic trajectories have the same behavior attribute characteristics as the user’s real trajectory.Experimental results show that this scheme can efficiently generate sets of fake locations for users and effectively confuse the user’s real trajectory with the synthetic trajectory generated by fake locations.