RcDT: Privacy Preservation Based on R-Constrained Dummy Trajectory in Mobile Social Networks

The boom of mobile devices and location-based services (LBSs) greatly enriches the mobile social network (MSN) applications, which bring convenience to our daily life and, meanwhile, raise serious privacy concerns due to the potential disclosure risk of location privacy. Besides the single-location privacy, trajectory privacy is another important type for location privacy leakage. In this paper, focusing on the trajectory privacy preservation in MSNs, we propose a privacy preservation scheme based on the radius-constrained dummy trajectory (RcDT) in MSNs. Particularly, by constraining the generated circular range with radius
$R$
for the location where a user sends LBS requests, we present the radius-constrained dummy location (RcDL) algorithm to generate the dummy location set of the user’s real location. Furthermore, based on the generated dummy locations, we put forward the RcDT algorithm to generate the dummy trajectory set that has higher similarity to the real trajectory comprehensively considering the constraints of both the single-location exposure risk and trajectory exposure risk. Thus, the user’s trajectory privacy preservation in MSNs is enhanced since the possibility of identifying users’ real trajectories and malicious attacks are reduced. The simulation results demonstrate that our RcDT scheme can have better performance and privacy degree than the existing methods.