Search for a command to run...
The market for Electric Vehicles (EVs) has expanded tremendously as seen in the recent Conference of the Parties 27 (COP27) held at Sharm El Sheikh, Egypt in November 2022. This needs the creation of an ecosystem that is user-friendly and secure. Internet-connected Electric Vehicle Charging Stations (EVCSs) provide a rich user experience and add-on services. Eventually, the EVCSs are connected to a management system, which is the Electric Vehicle Charging Station Management System (EVCSMS). Attacking the EVCS ecosystem remotely via cyberattacks is rising at the same rate as physical attacks and vandalism happening on the physical EVCSs. The cyberattack is more severe than the physical attack as it may affect thousands of EVCSs at the same time. Intrusion Detection is vital in defending against diverse types of attacks and unauthorized activities. Fundamentally, the Intrusion Detection System’s (IDS) problem is a classification problem. The IDS tries to determine if each traffic stream is legitimate or malicious, that is, binary classification. Furthermore, the IDS can identify the type of malicious traffic, which is called multiclass classification. In this paper, we address IoT security issues in EVCS by using different machine learning techniques and using the native IoT dataset to discover fraudulent traffic in EVCSs, which has not been performed in any previous research. We also compare different machine learning classifier algorithms for detecting Distributed Denial of Service (DDoS) attacks in the EVCS network environment. A typical Internet of Things (IoT) dataset obtained from actual IoT traffic is used in the paper. We compare classification algorithms that are placed in line with the traffic and contain DDoS attacks targeting the EVCS network. The results obtained from this research improve the stability of the EVCS system and significantly reduce the number of cyberattacks that could disrupt the daily life activities associated with the EVCS ecosystem.