Secure localization of land vehicles under GPS spoofing attack with decomposition Kalman filter

This study presents a secure vehicle localization framework designed to maintain accuracy under GPS spoofing attacks. The framework integrates a spoofing detection mechanism operating in three adaptive modes and employs refined dynamic, measurement, and attack models. A decomposition-based Kalman filter is utilized to develop a fusion algorithm for robust state estimation. Simulation and field experiments confirm that the proposed approach achieves high localization accuracy and resilience against spoofing in complex urban environments.