A Machine Learning-Based Intrusion Detection System for IoT Electric Vehicle Charging Stations (EVCSs)

The demand for electric vehicles (EVs) is growing rapidly. This requires an ecosystem that meets the user's needs while preserving security. The rich data obtained from electric vehicle stations are powered by the Internet of Things (IoT) ecosystem. This is achieved through us of electric vehicle charging station management systems (EVCSMSs). However, the risks associated with cyber-attacks on IoT systems are also increasing at the same pace. To help in finding malicious traffic, intrusion detection systems (IDSs) play a vital role in traditional IT systems. This paper proposes a classifier algorithm for detecting malicious traffic in the IoT environment using machine learning. The proposed system uses a real IoT dataset derived from real IoT traffic. Multiple classifying algorithms are evaluated. Results were obtained on both binary and multiclass traffic models. Using the proposed algorithm in the IoT-based IDS engine that serves electric vehicle charging stations will bring stability and eliminate a substantial number of cyberattacks that may disturb day-to-day life activities.