Publication | Open Access
Eviction of Misbehaving and Faulty Nodes in Vehicular Networks
400
Citations
32
References
2007
Year
Vehicle CommunicationInternet Of VehicleEngineeringInformation SecurityNetwork AnalysisVehicular NetworksHardware SecurityNode Certificate RevocationSystems EngineeringVehicle NetworkInternet Of ThingsFaulty NodesNetwork SecurityConvincing InstantiationAutomotive SecurityData SecurityCryptographyNetwork ScienceSurvivable Network
Vehicular networks are emerging as a mobile networking technology, but security is critical and misbehaving or faulty nodes pose a hard problem, especially given the lack of ubiquitous infrastructure for timely revocation. This paper tackles the challenge of detecting and preventing disruptive nodes in vehicular networks. The authors propose a framework of protocols that identify, locally contain, and evict misbehaving or faulty nodes, tailored to vehicular network characteristics and analytically evaluated. Results demonstrate that the distributed containment and eviction approach is feasible and provides sufficient robustness.
Vehicular networks (VNs) are emerging, among civilian applications, as a convincing instantiation of the mobile networking technology. However, security is a critical factor and a significant challenge to be met. Misbehaving or faulty network nodes have to be detected and prevented from disrupting network operation, a problem particularly hard to address in the life-critical VN environment. Existing networks rely mainly on node certificate revocation for attacker eviction, but the lack of an omnipresent infrastructure in VNs may unacceptably delay the retrieval of the most recent and relevant revocation information; this will especially be the case in the early deployment stages of such a highly volatile and large-scale system. In this paper, we address this specific problem. We propose protocols, as components of a framework, for the identification and local containment of misbehaving or faulty nodes, and then for their eviction from the system. We tailor our design to the VN characteristics and analyze our system. Our results show that the distributed approach to contain nodes and contribute to their eviction is efficiently feasible and achieves a sufficient level of robustness.
| Year | Citations | |
|---|---|---|
Page 1
Page 1