부산대학교 도서관

Vehicle-to-Everything (V2X) is an emergent technol-ogy for enhancing traffic efficiency, road safety and autonomous driving. Vehicles interconnected with their prevalent wireless environment are prone to various security threats that might affect traffic and life safety immensely. Jamming attacks, a legacy and dated problem, still persists much to the havoc of V2X communications. The following paper proposes a framework for jammer detection adapted to V2X communications scenario. A Generalized Dynamic Bayesian network is used to learn the V2X signal environment in a statistical manner. Subsequently, a Modified Markov Jump Particle filter (M-MJPF) is used for signal predictions where the innovations in the observed signal versus the predicted signal enable our framework to detect the jammer. Simulation results highlight the efficacy and accuracy of our approach in V2X jammer detection.