Research On Privacy Protection Models And Algorithms Based On Location Services

With the rapid popularization of smart phones, as well as the development of wireless communication technology, mobile users can enjoy location-based services anytime and anywhere. While enjoying the convenience brought by the location-based service, users also face the threats of location privacy disclosure. At present, the location privacy protection methods can be classified into adopting one of two main architectures, trusted third party architecture and P2P self-organizing network architecture. However, the third party tends to become a bottleneck of system performance and P2P architecture has to endure low anonymization success rates and suffers from kinds of attacks. Aimed at improving the ability to resist attacks in P2P architectures and enhancing the efficiency of location privacy protection methods, this paper has conducted research of the related problems about the location privacy protection, the main research works and contributions of this thesis are summarized as follows:(1) The architecture and mainstream methods of location privacy protection are summarized, and the models of attacker in location service are introduced in detail. Then, the existing protected methods and protocols of the location privacy protection are analyzed.(2) Since most of existing methods are based on the assumption that all the users trust each other, we present untrusted nearest neighborhood (UNNC) solution in the P2P architecture where no trust relationships are assumed. The main idea is to let users to share the unique VERF code which is posted by a third-party validator instead of their exact locations information so as to defend the attacks. The simulations show that the UNNC algorithm can effectively resist attacks from malicious users.(3) Since the cloaking region which generated by PrivacyGrid algorithm is larger than need, we propose improved algorithm which will generate sub cloaking regions and adopt a hybrid architecture to balance the work load between the third party server and the mobile clients. Hybrid structure chooses to use third party architecture or P2P architecture by users density, and applies improved algorithm when choose the third party architecture. The simulations show that the improved algorithm can reduce the computing pressure, and the hybrid structure can balance the work pressure of the central server and the mobile clients very well.