Towards a Lightweight Intrusion Detection Framework for In-Vehicle Networks

With the emergence of networked devices, from Internet Things (IoT) nodes and cellular phones to vehicles connected Internet, there has been an ever-growing expansion attack surfaces in Vehicles (IoV). In past decade, a rapid growth automotive industry as network-enabled electronic devices are now integral parts vehicular ecosystems. These include development automobile technologies, namely, Connected Autonomous (CAV) electric vehicles. Attacks on IoV may lead malfunctioning Electronic Control Unit (ECU), brakes, control steering issues, door lock issues that can be fatal CAV. To mitigate these risks, is need for lightweight model identify attacks systems. this article, efficient Intrusion Detection System (IDS) developed detect anomalies system. The dataset used study In-Vehicle Network (IVN) communication protocol, i.e., Area (CAN) generated real-time environment. classifies different types into reconnaissance, Denial Service (DoS), fuzzing attacks. Experimentation with performance metrics accuracy, precision, recall, F-1 score compared across variety classification models. results demonstrate proposed outperforms other