Signal Characteristic Analysis and Anomaly Detection for GPS Spoofing Mitigation

Abstract

This research addresses the critical issue of GPS spoofing attacks in Vehicular Ad-hoc Networks (VANETs), which pose significant threats to the safety and reliability of intelligent transportation systems. The study investigates effective strategies for detecting, tolerating, and managing these attacks, focusing on the unique challenges presented by the dynamic and distributed nature of VANETs. A novel hybrid machine learning approach, combining Bayesian Networks and a Watchdog Model, was developed to enhance anomaly detection in real-time GPS and network data. The methodology also incorporated advanced cryptographic techniques, signal characteristic analysis, and network intrusion detection systems to create a multi-layered defense mechanism. Experimental results, including data from a live spoofing event, demonstrated the effectiveness of the proposed methodology in accurately identifying and mitigating spoofing attacks, even in complex scenarios. The research findings have significant implications for enhancing the resilience of GPS-enabled VANETs against spoofing threats, paving the way for safer and more efficient transportation systems. Future research directions include refining the measurement models, exploring additional data sources, and developing more sophisticated attack scenarios to further strengthen the security of VANETs.