Joint Provisioning of QoS and Security in IoD Networks: Classical Optimization Meets AI

Abstract

The emerging unmanned aerial vehicle (UAV)-based communication networks, with the tight integration of the Internet of Things (IoT), are rapidly gaining prominence as the emerging Internet of Drones (IoD) paradigm. While IoD appears as an agile network to serve both mobile users and massive machine-to-machine communication and IoT traffic, it confronts sharply increasing security and privacy concerns due to the exponentially rising “untrusted” links and zero-day attacks in 5G and beyond networks. While security provisioning is important, its strict enforcement is known to significantly impact the service quality of the users. In this article, by discussing the tradeoff between tightly coupled provisioning of the quality of service and security performance of users, we first formulate an optimization problem to maximize the security levels subject to constraints specific to the considered IoD system. Then we demonstrate that the formulated problem is computationally hard. We describe existing methodologies that may be readily applicable to solve the problem, which may, however, not yield any solution even if a solution exists. To address this problem, we propose a simple but effective adaptive boosting-based method to maximize the security levels without significantly sacrificing network performance. The effectiveness of our proposed method is demonstrated and compared with existing methods using computer-based simulations.