Research On Privacy Preservation For Continuous Queries In Location-based Services

Rapid advancement in wireless communication and mobile positioning technologies facilitate the widespread use of location-based services(LBS).By using LBS,mobile users can obtain their required information at anytime and anywhere.Although these services bring convenience to people,they also threaten their privacy severely.Privacy concern in location-based services can be classified into two types: location privacy and query privacy.The most widely used models to address these issues are k-anonymity and l-diversity,respectively.However,most of these approaches are only limited to independent queries.When a user issues continuous queries,directly applying these approaches may lead to privacy disclosure severely.So,our paper focuses on the approaches of privacy preservation in continuous queries.In this paper,we first analyze the issues of location privacy and query privacy in location-based services and the related approaches to address these issues in detail.For continuous queries,we find out that simply using the number of users and queries' categories to evaluate the approaches' privacy protection level is not enough.So,based on users' footprints and historical queries information,we propose region entropy and query entropy to measure the privacy protection level for users' location privacy and query privacy,respectively.To calculate the region entropy and query entropy conveniently,we divide the whole terrain into grid cells,and then we separately design footprint table and query table to save the footprints and historical query information.For locating and retrieving users efficiently,we index all users by using Hilbert curve and B+-tree structure,and a new transformation formula between users' Hilbert values and grid cells' Hilbert values is presented.Then we propose Hilbert-based continuous cloaking and continuous obfuscation algorithms to protect users' location privacy and query privacy in continuous queries comprehensively.At last,we design variances of experiments to evaluate the efficiency and effectiveness of our proposed approaches.