Toward a Secure Drone System: Flying With Real-Time Homomorphic Authenticated Encryption

Abstract

Controlling or accessing remotely has become a prevalent form of operating numerous types of platforms and infrastructure. An exploding number of vehicles such as drones or cars, in particular, are being controlled wirelessly or connected through networks. This has brought unanimous concern that today's networked vehicle systems are vulnerable to attacks and the results could be fatal. Unfortunately, in contrast to active investigation on the security of the vehicles themselves, sensors, or communication channels, existing approaches for these real-time, safety-critical systems do not take controllers into enough consideration. In order to protect the controller that performs the arithmetic operations using sensor measurements and generates command signals, we adopt homomorphic cryptography for the controller. It removes risks associated with the management of the secret key inside the controller, by eliminating the need to encrypt and decrypt the data for the mathematical operation within the controller. Specifically, we propose an efficient linearly homomorphic authenticated encryption (LinHAE) scheme for the ground control center of a multi-rotor drone, in a manner that enables real-time operation for safe autonomous flight. To facilitate the linear scheme, we design the ground controller targeted to allow state update using additions and multiplications by a system-specific constant. The proposed LinHAE guarantees the security against eavesdropping and forgery attacks, unlike homomorphic encryption alone that does not provide means to check whether the received signal at the drone side is authentic or compromised. We introduce a LinHAE with security and computational tractability, and describe how it can fit into the standard architecture for drone systems and how the specific controller is implemented. Building on these ingredients, we report the first successful operation of a multi-rotor flying robot that autonomously flies under the ground controller with real-time homomorphic authenticated encryption.

Keywords

• Cryptography
• encryption
• cyber-physical systems
• control design
• unmanned aerial vehicles