A Lightweight and Efficient Authentication Scheme for the Internet of Drone Based on Cancelable Biometrics

ABSTRACT: Unmanned aerial vehicles (UAVs is also known as drones) have significant applications in smart cities, and the information exchange between UAVs and the control server (CS) is conducted through wireless communication channels, which are susceptible to various security risks, such as network attacks and drone capture. To ensure the security and integrity of information in the Internet of Drones (IoD), identity authentication and key agreement protocols can be designed for protection. However, due to the unique characteristics of IoD, such as the extremely high mobility of drones in real scenarios and the resource constraints of drones, there is a need to meet the requirements for lightweight protocols. This paper proposes a strategy that uses cancelable biometric features to protect the biometric features of users during the authentication process. The method combines Fast Fourier Transform, Gaussian random projections, Position-Sensitive Hashing, fuzzy extractors, and Physical Unclonable Functions (PUF), meeting the security and lightweight needs of IoD authentication protocols. We use the Real-or-Random (ROR) model and the Avispa simulation tool to prove that our protocol is secure. Through comparative research, the proposed cancelable method has higher matching efficiency and better unlinkability, and our protocol offers higher security and faster computational efficiency.