Internet of Vehicle (IoV) Cyber Attack Detection using Machine Learning Techniques

In the modern century, sophisticated interconnected technologies, also known as the Internet of Things (IoT), evolve quickly and become more adaptable and intelligent, influencing our existence, occupation, transportation, and interaction. The capacities of IoT equipment render the traditional transportation framework a progressively more sophisticated transportation infrastructure, designated as the IoV (Internet of Vehicles). In IoV interactions, the controller area network (CAN) protocol manages vehicle sensors. Due to a lack of communication confidentiality and authentication in the IoV CAN protocol, its communications are exposed to data vulnerability, fuzzy attacks, spoofing, denial of service attacks, and several other significant security issues. They trigger the failure of the electronic control units (ECUs) of these IoVs, which could cause anomalies with breaking, locking, steering, ventilation, and other systems. In the worst cases, this could even be fatal. The proposed research's primary goal is to improve the broad cybersecurity concerns in IoVs using machine learning (ML) methodologies and ensure consumer safety on the street. By incorporating the CICIoV2024 dataset and cutting-edge ML techniques, the proposed solution is able to determine assault over IoVs with the highest possible efficiency, i.e., 100 percent.