Searching for Scalable Networks in Unmanned Aerial Vehicle Infrastructure Using Spatio-Attack Course-of-Action

Abstract

Unmanned aerial vehicles are increasingly being applied to various applications for a variety purposes, such as delivery, communication relay, mapping, and surveillance services. Through these, it is possible to provide flexible and stable network services. Unmanned aerial vehicles perform a wide range of tasks using Internet-of-Things technology, which needs Internet access. These internet connections, however, make it more possible for attackers to execute various security attacks on unmanned aerial vehicles. Therefore, it is crucial to identify the attack behavior of the adversary, which is called “course-of-action”, to preserve security in the unmanned aerial vehicle infrastructure. Based on learned data, the existing course-of-action method has the drawback of not functioning on various networks. As a result, in this paper, we propose a novel heuristic search-based algorithm to apply to various unmanned aerial vehicle infrastructures. The algorithm can build the optimal heuristic functions in various unmanned aerial vehicle network environments to explore the attack course-of-action and design the optimal attack paths to maximize total reward. Applying the proposed algorithm in two unmanned aerial vehicle network scenarios allowed us to confirm that the best attack path is well established.