- https://doi.org/10.22266/ijies2026.0228.39
Machine Learning-based DDoS Attack Mitigation in EV Charging Station Authentication Systems
The growing adoption of electric vehicles (EVs) has heightened the importance of securing charging station authentication systems, particularly against distributed denial of-service (DDoS) attacks that threaten service availability.This study proposes a machine learning-based framework to detect and mitigate DDoS attacks targeting EV charging authentication processes.Using the domain-specific CICEV2023 dataset, we evaluate several machines learning models, including Random Forest, Logistic Regression, K-Nearest Neighbors (KNN), XGBoost, CatBoost, Multilayer Perceptron (MLP), and a Hybrid Voting Classifier.Feature selection techniques are applied to improve both accuracy and computational efficiency.Experimental results indicate that the MLP model achieves the highest classification accuracy of 96.2%, with superior recall and F1-scores.Feature selection significantly enhances model performance and reduces prediction latency, making the solution viable for real-time IoT-based deployments.The results indicate that machine learning models, particularly deep learning methods, are good at keeping EV charging station authentication systems safe from evolving cyber threats and are also appropriate for real-time use and applications with limited resources on the Internet of Things.
- # Distributed Denial Of-service
- # Evolving Cyber Threats
- # Machine Learning Models
- # Hybrid Voting
- # Electric Vehicles Charging
- # Superior Recall
- # Deep Learning Methods
- # Internet Of Things
- # Multilayer Perceptron
- # K-Nearest Neighbors