Detection and localization of sybil nodes in VANETs

Abstract

Sybil attacks have been regarded as a serious security threat to ad hoc networks and sensor networks. They may also impair the potential applications of VANETs (Vehicular Ad hoc Networks) by creating an illusion of traffic congestion. In this paper, we present a lightweight security scheme for detecting and localizing Sybil nodes in VANETs, based on statistic analysis of signal strength distribution. Our scheme is a distributed and localized approach, in which each vehicle on a road can perform the detection of potential Sybil vehicles nearby by verifying their claimed positions. We first introduce a basic signal-strength-based position verification scheme. However, the basic scheme proves to be inaccurate and vulnerable to spoof attacks. In order to compensate for the weaknesses of the basic scheme, we propose a technique to prevent Sybil nodes from covering up for each other. In this technique, traffic patterns and support from roadside base stations are used to our advantage. We, then, propose two statistic algorithms to enhance the accuracy of position verification. The algorithms can detect potential Sybil attacks by observing the signal strength distribution of a suspect node over a period of time. The statistic nature of our algorithms significantly reduces the verification error rate. Finally, we conduct simulations to explore the feasibility of our scheme.