Research On Privacy Protection And Query Authentication Strategies Of Location Based Services

With the popularity of mobile positioning and communication technologies,many location-based services associated with mobile objects can be supported by analyzing and mining real-time location or trajectory data of users.Existing location services mainly include user-initiated rea-ltime location-based spatial query services and third-party provided commercial location services based on user trajectory data or continuously issued location information.While the location information of users can bring great benefits to people,information such as location data and personal privacy also face serious leakage problems.At the same time,how to ensure high-quality location services and correct spatial query results are also issues that users value.Our research aims at the spatial query service such as group nearest neighbor query(GNN query),and fully considers the privacy protection of user location data and the valid verification of the correctness of spatial query results.In order to solve the problem of privacy protection of user location in GNN queries,we propose two effective privacy protection algorithms,namely geometric centroid method and encrypted centroid method.In order to solve the problem of verifying the correctness of spatial query results in GNN queries,we propose query processing and result verification based on Voronoi diagram.The algorithm ensures the privacy protection of the user location and the correctness of the query result,reduces the overhead of storage,communication and calculation as much as possible,and can effectively resist various privacy attack models.In the experimental part,through comparison with the three existing related algorithms,the results prove that this algorithm not only fully considers privacy protection and query verification,but also has more reasonable resource overhead.For the location service that providing user trajectory data or location information to third parties for various commercial purposes,this paper proposes an effective personalized privacy protection method in the offline trajectory data publishing scenario.The algorithm can achieve a good balance between user-defined privacy requirements and data availability.The main idea of the algorithm is as follows.First,the semantic attributes of all sampling points on the trajectory are marked,and the corresponding semantic tree is established.Then,all the sensitive stop points on the user's trajectory are extracted,and then in consideration of the user's speed and avoidance of reverse-mutation,different strategies are adopted to select a suitable user interest point for different types of sensitive stop points.Finally,we release the reconstructed user trajectory to a third party.In addition,in order to make our method more in line with the actual situation,the algorithm avoids the situation where the reconstructed trajectory segment has a location mutation or goes through a space obstacle.The experimental part adopts the trajectory average identification possibility,trajectory semantic consistency and trajectory shape similarity as the evaluation criteria,then carries out a comprehensive evaluation of the algorithm proposed in this paper.The experimental results show that our method can improve the trajectory availability of users as much as possible,and effectively achieve different trajectory privacy requirements.