GPS Spoofing Detection on Autonomous Vehicles with XGBoost

Global Positioning System (GPS) spoofing presents a significant threat to the reliability and security of global navigation systems by impacting various critical sectors such as military, transportation, communication, and finance. Traditional methods for detecting GPS spoofing often fall short in addressing sophisticated spoofing attacks. In our work, we investigate the application of machine learning techniques to enhance the detection of GPS spoofing. We aim to identify patterns and anomalies in GPS signals that indicate spoofing attempts by leveraging data-driven approaches. Our proposed method involves training machine learning models on a dataset comprising both legitimate and spoofed GPS signals. The results obtained from this work demonstrate the effectiveness of these models in accurately detecting spoofing incidents, particularly with t he tree-based machine learning model XGBoost. This research underscores the potential of machine learning to provide robust, real-time spoofing detection, thereby enhancing the resilience of GPS-dependent systems against malicious attacks, with XGBoost standing out for its high accuracy.