A location privacy preserving solution to resist passive and active attacks in VANET

Abstract

Existing location privacy-preserving methods, without a trusted third party, cannot resist conspiracy attacks and active attacks. This paper proposes a novel solution for location based service (LBS) in vehicular ad hoc network (VANET). Firstly, the relationship among anonymity degree, expected company area and vehicle density is discussed. Then, a companion set V is set up by k neighbor vehicles. Based on secure multi-party computation, each vehicle in V can compute the centroid, not revealing its location to each other. The centroid as a cloaking location is sent to LBS provider (P) and P returns a point of interest (POI). Due to a distributed secret sharing structure, P cannot obtain the positions of non-complicity vehicles by colluding with multiple internal vehicles. To detect fake data from dishonest vehicles, zero knowledge proof is adopted. Comparing with other related methods, our solution can resist passive and active attacks from internal and external nodes. It provides strong privacy protection for LBS in VANET.