Application of non-cooperative game theory in analyzing a potential DoS on autonomous vehicular network

Autonomous systems are soon expected to integrate into our lives as home assistants, delivery drones, and driverless cars. Given the importance of cybersecurity of these systems, game-theoretic solutions provide a mathematical modeling of the interaction between malicious users and the system. This work introduces a novel framework for analyzing potential denial of service (DoS) attacks on autonomous vehicular networks by integrating a non-cooperative, non-zero-sum game model with a comprehensive simulation environment (VEINS/OMNET++/SUMO). Our key contribution lies in the systematic quantification of attacker and defender payoffs based on realistic network metrics, enabling the identification of optimal strategies across 21 distinct attack/defense scenarios derived from tunable network parameters. We specifically emphasize the model’s ability to provide a predictive, all-scenario evaluation to aid system designers in preparing for real-time cyber attacks on driverless cars.