Privacy-Aware VANET Security: Putting Data-Centric Misbehavior and Sybil Attack Detection Schemes into Practice

Abstract

The past decade has witnessed a growing interest in VANET (Vehicular Ad Hoc NETwork) and its myriad potential applications. Nevertheless, despite the surge in VANET research, security and privacy issues have been the root cause of impeded momentum in VANET deployment. In this paper we focus on misbehavior and Sybil attacks from VANET standpoint. With intrusion capabilities in hand, malicious users in VANET can inject false information and launch Sybil attack. Sybil attack refers to pretending one physical node to be many and in worst case almost all kinds of attacks can be launched in the presence of Sybil attack. Misbehavior in VANET can be categorized as a sub-effect of Sybil attack where a malicious vehicular node(s) spoof legitimate identities. There are two main strategies for avoiding misbehavior in VANET; Entity-centric strategies that focus on the revocation of misbehaving nodes by revocation authorities. On the other hand, Data-centric approach mainly focuses on the soundness of information rather than the source of information. We cover both strategies where decision on which strategy to be used, is taken on the basis of traffic situation. In a dense traffic regime, we propose SADS (Sybil Attack Detection Scheme) whereas in sparse traffic regime, we propose LMDS (Location-Based Misbehavior Detection Scheme). Our proposed schemes leverage position verification of the immediate source of warning message. Furthermore, we guarantee security and privacy (conditional anonymity) for both beacons and warning messages.