Research On Location Privacy Protection Techniques For Location-based Service

Location-Based Service (LBS) is more and more popular and widely used in people's daily life as the rapid development of wireless positioning technologies and mobile Internet technologies. LBS is a location based customized service. Examples of such services include traffic guidance, POI requests, vehicle tracing, emergency call and etc. To get the precise answers of these services, a user has to provide his/her exact location to the LBS. Since the location is very sensitive to users'privacy, users are at the risk of exposing location privacy when enjoying the convenience of LBS. To achieve a tradeoff between the quality of service and location privacy protection has become a research hotspot. Our contributions in this paper are outlined as follows.Firstly, to balance the contradiction among service quality, privacy and computation complexity, a novel k-anonimity scheme based on service similarity is proposed. With server-central framework, TTP applies the location service similarity to creat cloaked query region. The scheme eliminates the blindness of the traditional construction method based on random strategy. It not only increases the construction efficiency of cloaked query region and the quality of service, but also improves the query efficiency of the LBS server.Secondly, to address the issue that the location k-anonymity scheme has proved to be vulnerable to expose the user's identity in continuous LBS, a novel location cloaking scheme based-on/-queries is proposed. In the proposed scheme, we insert fake queries randomly into real user's query. This scheme achieves the privacy protection in single query by generalized the query theme. Moreover, it equalizes the probability of each query at location server side, disables the continuity of queries at continuous time series, eliminates the integrality of trajectories and enhances the ability of the privacy-preserving scheme against continuous attack effectively. Besides, the partitioning algorithm and queries clustering algorithm are introduced to reduce computation and communication overhead further.Finally, to protect trajectory privacy in continuous queries, a trajectory l-anonymity protection scheme based on road networks is proposed. This scheme depersonalizes a user's trajectory by generating similar trajectories on the road map. Instead of reporting each location to LBS server, this scheme sends a cloaked query region that contains l different trajectories. It means that the attackers cannot distingwhish the real trajectory from the similar l-trajectories. Furthermore, we add the fake ID mapping based-on k-anonymity to improve the privacy. The scheme can achieve trajectory protection of l-diversity with a minimized cloaking query region.