RLAP: Revocable Lightweight Authentication Protocol for Adaptive Security in Internet of Drones

Abstract

Internet of Drones (IoD) has become increasingly important to support collaborative Unmanned Aerial Vehicles (UAV) services such as logistics, surveillance, and disaster recovery. Yet, current authentication mechanisms for IoD are typically based on static credential models and do not support revocation in real-time, thereby exposing such environments to insider threats, and unauthorized access. In response, in this paper, propose Revocable Lightweight Authentication Protocol (RLAP), a scalable and decentralized authentication which introduces trusted units to identify fellow drones using ephemeral elliptic curve keys and pseudo identifiers. RLAP leverages a Bloom filter based revocation strategy for constant detection of compromised nodes without requiring centralized storage or synchronized clocks. Our protocol has low computational and communication overhead and can achieve stateless mutual authentication. Analysis against the security requirements shows RLAP is secure against impersonation, replay, and verifier compromise, and the experimental results reveal the computational cost of RLAP is reduced by 57% and message size is decreased by more than 60% in comparison to the existing schemes. RLAP is well-suited to safe and adaptive communication on large-scale, resource limited UAV swarms.