TinyML-Enabled Intrusion Detection for Securing Electric Vehicle Supply Equipment (EVSE)

Electric Vehicles (EVs) and Electric Vehicle Supply Equipments (EVSEs) are vulnerable to cybersecurity threats like malware injections, unauthorized access, denial-of-service (DoS) attacks, distributed denial-of-service (DDoS) attacks, and man-in-the-middle (MITM) attacks, which can compromise their functionality and security. While Machine Learning (ML) effectively addresses cybersecurity threats, implementing traditional ML solutions in EVSE is challenging due to resource limitations, requiring lightweight and efficient solution to ensure effective threat detection and mitigation. Tiny Machine Learning (TinyML) is an emerging solution that enables running ML models on resourceconstrained devices, making it ideal for implementing Intrusion Detection Systems (IDS) in EVSE charger units. This paper explores the potential of TinyML as an effective solution for enhancing EVSE cybersecurity. The proposed TinyML model is evaluated using the Hardware Performance Counter (HPC) and Network data sources of CICEVSE2024 dataset, achieving accuracies of 98.72% and 99.84%, F1-scores of 98.65% and 99.83%, recall scores of 98.72% and 99.84%, and precision scores of 98.88% and 99.84%, respectively.