​

​Secure Vehicle-to-Everything (V2X) communication systems are essential for ensuring safe transportation in civilian applications, such as smart cities and military operations. This paper introduces S2VC, a secure smart vehicular communication solution that addresses everyday scenarios. S2VC analyzes existing vulnerabilities, implements various cyber-attacks (both internal and external), assesses their impacts, and deploys security measures to prevent and detect such attacks. To achieve this, testbeds using robot cars and virtual roads are constructed to simulate real-life scenarios, allowing different security attacks and defence solutions to be applied. A new, lightweight protocol has been developed in this research to facilitate communication between the cars. Additionally, traffic data from all vehicles in these scenarios is collected to create specialized datasets. These datasets are then used to train Artificial Intelligence (AI) algorithms to develop intelligent models capable of predicting and classifying different types of attacks. In this paper, we have assessed the performance of one of the specialized datasets labeled with denial of service (DoS) attacks. We tested this dataset using five machine learning (ML) algorithms. They all achieved excellent detection accuracy, ranging from 99.2% to 100%.

Presented by: Iman Almomani