Rogue Drone Detection: A Machine Learning Approach

Abstract

The emerging, practical and observed issue of how to detect rogue drones that carry terrestrial user equipment (UEs) on mobile networks is addressed in this paper. This issue has drawn much attention since the rogue drones may generate excessive interference to mobile networks and may not be allowed by regulations in some regions. In this paper, we propose a novel machine learning approach to identify the rogue drones in mobile networks based on radio measurements. We apply two classification machine learning models, Logistic Regression, and Decision Tree, using features from radio measurements to identify the rogue drones. We find that for high altitudes the proposed machine learning solutions can yield high rogue drone detection rate while not mis-classifying regular ground based UEs as rogue drone UEs. The detection accuracy however degrades at low altitudes.