LSR: Mitigating Zero-Day Sybil Vulnerability in Privacy-Preserving Vehicular Peer-to-Peer Networks

Abstract

Privacy-preserving Vehicular Peer-to-Peer Network (VPNET) is particularly vulnerable to sybil attack, where a malicious vehicle can abuse its multiple unlinkable pseudo-ids to pretend multiple and distinct vehicles in the network. To make the matter even worse, due to the privacy-preserving network environment, zero-day sybil vulnerability is hard to defend against, i.e., a vehicle cannot locally detect a sybil attacker before the attacker is formally revoked. In this paper, aiming at mitigating zero-day sybil vulnerability in privacy-preserving VPNET, we propose an efficient Local Sybil Resistance scheme, called LSR, to locally detect sybil attack. Especially, in the proposed LSR scheme, if a vehicle never signs an event more than once, the signatures it signed cannot be linked, and its privacy can be well protected. However, if a vehicle signs two or more signatures on the same event, any vehicle can easily link these signatures and thus detect a sybil attack locally. Moreover, with two-layer/multi-layer reporting, a sybil attack can be quickly reported to a trusted authority (TA) for tracking the sybil attacker's real identity and making global revocation. Detailed security analysis demonstrates that the proposed LSR scheme can enhance the security of a privacy-preserving VPNET, such as locally detecting sybil attack, preventing a sybil attacker's future attacks before its being revoked by TA, et al. In addition, performance evaluation via extensive simulations also confirms the high effectiveness of the proposed LSR scheme.