Modeling Roadside Attacker Behavior in VANETs

Abstract

Communication using VANETs is commonly seen as the next milestone for improving traffic safety. Vehicles will be enabled to exchange any kind of information that helps to detect and mitigate dangerous situations. Security research in the past years has shown that VANETs are endangered by a plethora of severe security risk.Subject of this work is the modeling of attackers that target active safety applications in VANETs. Through a risk analysis, this work identifies assets, threats and potential attacks in inter-vehicle communication. The risk analysis shows that the most serious threat arises from a quasi-stationary (road-side) attacker that distributed forged warning messages. This attacker is discussed more deeply. We show the degrees of freedom that are available for position forging and find thereby two attacks that demand attention: single position forging having low effort compared to sophisticated movement path forging having a potentially high influence on road traffic safety. I. INTRODUCTIONVehicular Ad-Hoc Networks (VANETs) describe the technology of direct communication among vehicles themselves as well as among vehicles and (roadside) infrastructure. They enable vehicles to exchange information in order to increase vehicle passenger safety and road safety, traffic efficiency and driver convenience. Such information are enhancing autonomous in-vehicle safety systems as well as they are informing the driver of relevant events where his reaction or attention is required. Thus, the information received by a vehicle must be secure and reliable, meaning that information security is a crucial part of such a system. Information security and hence drivers' safety is endangered once there is a vulnerability in the system attracting attackers who exploit the vulnerability according to their motivation. For example, the open system character of VANETs might motivate attackers to interfere with the system. This interference may even provoke wrong driving maneuvers leading to an accident in the worst case.Attacks on VANETs have been summarized generally in previous work [1]. Several solutions have been proposed to secure the system against a variety of these attacks. What is missing so far is an in-depth discussion and analysis of attackers and the modeling of attacker behavior to analyze and to help to improve the proposed security solutions.The early stage of development of VANETs does not allow for a meaningful attack analysis. Currently, there are still too many options with respect to protocols and applications. Modeling all possible attacker behaviors and attacks on these would be impossible. Therefore, we reduce the options by

Position forging attacks in Vehicular Ad Hoc Networks: Implementation, impact and detection

“…Roadside attacker behavior is discussed in detail by Leinmuller et al in [17]. This paper provides an overview of various position forging attacks.…”

A Novel Framework for Efficient Mobility Data Verification in Vehicular Ad-hoc Networks

“…A stationary roadside attacker is identified in [14] as the most threatening attack due to low complexity. As discussed in the attack on the GLOSA application, we additionally identify a moving attacker as rather threatening because its motivation may be higher compared to attackers that are not part of the road traffic.…”

Detecting misbehaviors in VANET with integrated root-cause analysis

“…The 1609.2 standard [3] also proposes the functionalities of a security layer in V2V communication. There have also been efforts regarding characterizing attacker behavior [13]. In a recent work, Sun and Fang [26] address the problem of defending against general misbehaviors in VANET.…”

“…The focus of this paper is to understand detection of misbehavior at the application layer. The related issues of understanding the attack and of security performance are considered in [13,14]. There exist several other works dealing with general considerations towards securing V2V systems [15][16][17][18][19]1,[20][21][22].…”