Implementing Gradient Boosting Techniques for Real-Time Attack Detection in Vehicular Networks

Several attacks pose serious dangers to vehicular networks (VN), crucial to creating smart transportation systems (ITS). Due to ITS, VN has become more important since it allows for smooth communication between infrastructure and cars. These networks are becoming more interdependent with smart transportation systems, which makes them more susceptible to assaults. By applying machine learning (ML) strategies, this paper seeks to strengthen VN security by making intrusion detection systems (IDS), applying ML models Gradient Boosting Machine (GBM), Adaptive Gradient Boosting (AdaBoost), Extreme Gradient Boosting (XGBoost), and Light Gradient Boosting Machine (LightGBM) to the ToN-IoT dataset for training and evaluation. The dataset contains normal and malicious network traffic data. These results show that XGBoost and LightGBM, two GB algorithms, are super effective in detecting attacks in VN. The research shows that these models can alert drivers of impending threats in real time, making vehicle communication systems much safer. More research is underway into testing these models in real-world vehicle networks under varying conditions and attack scenarios. Intelligent transportation systems rely on developing cutting-edge security solutions, which this research helps to provide.