Optimization of a Secure UAV-Based IoT: RF-Fingerprint Authentication and Resource Allocation

Abstract

Unmanned aerial vehicle (UAV) technology has developed to establish a mobile edge computing (MEC) network for the Internet of things (IoT). In the MEC network, users can reduce their latency by communicating and exchanging data with UAV-based edge servers. Security is a critical issue in a UAV-based IoT since the attackers who attempt to access the network may cause interference and influence the flight of the UAV. In this paper, we propose a lightweight RF fingerprinting recognition method in consideration of the limited computing power in UAVs, identifying unauthenticated attackers and refusing their access to IoT. Also, we propose a resource allocation scheme in the secure UAV-based MEC network. By establishing a non-convex resource optimization problem and decomposing it into a few tractable subproblems, we offer a numerical algorithm for the optimum resource allocation. The analysis results illustrate that our proposed method can reduce energy consumption and running time compared to its benchmark methods.