A Novel Protection Method of Continuous Location Sharing Based on Local Differential Privacy and Conditional Random Field

At present, the protection of user's location privacy (especially mobile user's location privacy) is widely concerned by academia and industry. Many experts and scholars have proposed several secure and efficient solutions to this problem. However, in this context, if a mobile user wants to obtain the services she/he wants, she/he needs to share her/his location information continuously with an untrusted third-party in user's locations. This will cause privacy and security issue. To tackle this problem, in this paper, we apply local differential privacy (LDP) in supporting continuous location sharing among mobile users. Firstly, we put forward a new idea of using conditional random field (CRF) in model user's mobility. Then, we combine delta-location set and epsilon-LDP and advance a mechanism to support continuous location sharing. Finally, we conduct experiments on real data set to evaluate our proposed mechanism. The results show that our mechanism is more effective than planar isotropic mechanism (PIM).