Research On Location Privacy Protection Technology Based On Differential Privacy

With the widespread popularity of mobile devices with positioning features,Location Based Services(LBS)have been widely applied.Users send personal location information to third-party providers of LBS service to obtain corresponding services.However,third-party providers of LBS service are untrusted.Attackers can infer the user's sensitive information from the user's exposed location data.This poses a serious threat to the user's personal privacy.How to protect location privacy has become the current Research hotspots.Existing location protection technologies,such as anonymity,mixed zones,and fake locations,all rely on existing known background knowledge.As the background knowledge of an attacker increases,location information may also be leaked.Differential privacy is a strong privacy protection model that has nothing to do with background knowledge.The privacy protection strength can be measured by privacy budget.Provides a new approach to location privacy protection.The main research work of this paper is as follows:1.Firstly,according to the access amount of geographic location,a user location privacy budget allocation method is proposed.This algorithm uses a grid to divide the user's access area,and based on the access amount in a certain period of time,the privacy budget of the corresponding area is allocated reasonably.Using the geographic indistinguishability mechanism,the location data is noisy and the interference location is generated.2.Aiming at the problem of low location nearest neighbor query hit rate caused by location disturbance,a combined incremental query algorithm is proposed to ensure that the supply space contains the possible offset region of the query target,which effectively reduces the offset band caused by the query target position.Coming error.The experimental results show that,while protecting the privacy of the location,the combined incremental query can effectively improve the query hit rate,which is suitable for the nearest neighbor query of the disturbed location.