SpoofGuard Framework: A Novel GPS Spoofing Detection and Mitigation Approach for Autonomous Vehicles Using Interval Kalman Filtering and Hybrid Signal Modeling

In autonomous vehicles, GPS is vital for navigation but vulnerable to spoofing attacks that compromise safety and system reliability. Traditional Kalman Filters (KFs) assume linear dynamics and Gaussian noise, which are inadequate for handling uncertainties and nonlinearities caused by GPS spoofing. This paper presents the SpoofGuard Framework, a GPS spoofing detection and mitigation method combining Interval Kalman Filtering (IKF) with hybrid signal modeling. Legitimate GPS signals are modeled using a Ricean distribution, while spoofed signals follow a Rayleigh distribution. The IKF adapts dynamically to both legitimate and spoofed signals, enabling precise position estimation and spoofed signal rejection. Unlike traditional KFs, which rely on point estimates, the IKF uses interval estimates, making it resilient to deviations and spoofing attempts. This approach enhances the security and reliability of autonomous vehicle navigation.