Research Area:  Vehicular Ad Hoc Networks
Critical event information dissemination has been proliferating on VANET allowing road safety via connected vehicular communications. Despite the prospect of promising applications in vehicular networks, it faces unresolved challenges that hold the capability to slow down network performance upon deployment, especially in terms of security. Particularly, insider attacks such as Blackhole attacks that are carried out against VANET systems can disrupt the networks’ average performance and prevent communication between vehicles entirely. Many state-of-the-art solutions have been proposed to detect and eliminate such nodes based on reputation systems and broadcast routing. However, if the network consists of multiple malicious nodes, the message dissemination could fail due to broadcast message tampering attack or packet dropping. In this study, we explore to answer the question of “can we improve the insider attacks mitigation in VANET by enhancing the trust in the network system so that the possibility of successful attacks can be reduced?”. To answer this question, in this paper, we present the blockchain-based decentralized trust score framework for the participating nodes to detect and blacklist insider attackers in VANET proactively. We propose a two-level detection system, in which at the first level, neighboring nodes calculate the trust individually. In the second level, a consortium blockchain-based system with authorized Road Side Units (RSUs) as validators, aggregate trust scores for vehicular nodes. Then, based on trust scores reported by the neighboring nodes, the blacklist node tables are dynamically modified. The experimental analysis shows that the proposed system is efficient and scalable in terms of the network’s practical size. Finally, we also present evidence that the proposed system improves the VANET performance by mitigating and blacklisting insider attack launching nodes.
Author(s) Name:  Sowmya Kudva,Shahriar Badsha,Shamik Sengupta,Hung La,Mohammed AtiquzzamanIbrahim Khalil
Journal name:  Journal of Parallel and Distributed Computing
Publisher name:  ELSEVIER
Volume Information:  Volume 152, June 2021, Pages 144-156
Paper Link:   https://www.sciencedirect.com/science/article/abs/pii/S0743731521000459