ARES: An anonymous region expanding system for location privacy preservation

Abstract

Generalization is a common strategy for privacy preservation, but in the process of generalization, anonymous users are usually selected by the initiator from nearby candidates, which makes the initiator often located at the center and easily predictable by the adversary. Moreover, a higher density of anonymous users in the anonymous region usually leads to excessive concentration, which can also be used to jeopardize the initiator’s location privacy. To address the above issues, based on the discretized distances among anonymous users, a new location privacy preservation system is proposed. In this system, the region where the initiator is located is divided into discrete areas according to the density of anonymous users, and then these areas are linked by the Hilbert curve in the order of increasing user distance. After that, the density of each area is stored in different nodes and establishes a n-level location quadtree. With this tree, the initiator can choose a suitable number of anonymous users in different areas according to discrete degree, and then generalize the initiator’s true location. In addition, as the initiator is no longer located at the center of the region, the proposed system can also effectively resist the attack of center selecting, identical PoI attack and density difference attack as well as some other types of attacks. Finally, an engineering application and theoretical analysis of the proposed system is given, and the proposed system is also compared with several similar systems.