USENIX Security '22 - An Experimental Study of GPS Spoofing and Takeover Attacks on UAVs
Harshad Sathaye, Northeastern University; Martin Strohmeier and Vincent Lenders, armasuisse; Aanjhan Ranganathan, Northeastern University
Today, there is limited knowledge about the behavior of UAVs under GPS spoofing attacks in a real-world environment, in particular considering the interplay between the UAV's software as well as other equipped navigation aids and vision sensors. This work aims to understand the feasibility and requirements of fully controlling a UAV's movements by spoofing GPS signals alone. We enumerate the challenges in accomplishing a complete UAV takeover through GPS spoofing and controlling it without crashing. We design and implement a Real-time GPS Signal Generator (RtGSG) that can be configured to generate any arbitrary trajectory and is capable of making changes to GPS signals in real-time through user input, e.g., using a keyboard or joystick. We evaluate RtGSG on popular commercial UAVs from DJI and Autel through over-the-air spoofing experiments in a controlled chamber. We explore generic and UAV-specific GPS spoofing strategies in order to best achieve complete maneuvering control (e.g., velocity and direction). This work highlights that, although COTS UAVs remain vulnerable to GPS spoofing attacks, a complete takeover and control of the UAV requires careful manipulation of the spoofing signals in real-time. Finally, we release our implementation to the scientific community for further research.
View the full USENIX Security '22 program at [ Ссылка ]
