An Approach to Mitigate Multiple Malicious Node Black Hole Attacks on VANETs

International audienceInter-vehicular communication is a growing trend among the motor vehicle industry, as part of theshift to the Intelligent Transportation System (ITS) paradigm and the emergence of autonomous vehicles. Thebenefits of using vehicular communication systems are many: traffic congestion can be abated leading toincreased route efficiency and fuel economy; road traffic accidents and fatalities can be prevented; and newservices are made available to both driver and passengers. Due to the important functions carried out by thesenetworks, it is essential that communication is secure and free from interference by malicious parties. But thevery nature of these networks (i.e. public and dynamically forming, over a shared medium) make themsusceptible to certain attack vectors. Black hole attacks are a discernible threat to vehicular communicationand the availability of Vehicular Ad Hoc Networks (VANETs). Previously, we proposed a solution to mitigatethese attacks in the case of a single malicious node. This countermeasure comprised a detection, accusation,and blacklisting scheme. We now extend that work to also account for cases where multiple malicious nodesare present in the same network. For cases where there is only one malicious node on multiple routes, we adda recursive element to our algorithm to handle additional detections during the accusation phase. Routescontaining a node with a pending accusation are also precluded from selection. Where multiple maliciousnodes exist on the same route, we adapt our former strategy to allow for multiple concurrent accusations andto find alternate routes to non-malicious nodes that need to be queried when verifying an accusation. Weprovide validation for our proposal using a simulation model built in NS-3. Results from simulation show theefficacy of our solution in detecting and preventing multiple malicious node black hole attacks in VANETs