EA-UAP: A Software-Engineered Adaptive and Energy-Aware Authentication Protocol for 5G-Enabled UAV Networks

Unmanned Aerial Vehicles (UAVs) deployed in 5G-powered Internet of Drones (IoD) systems need secure authentication solutions, that can achieve a high level of security assurance yet would not impose an excessive overhead on the onboard energy consumption. This work introduces EA-UAP, a software-engineered adaptive and energy-aware authentication mechanism suitable for 5G-connected UAV environments. The research input is a dynamic adaptive scheduling mechanism from full authentication, fast re-authentication and session extension modes with the consideration of two context mitigation parameters, Security Risk Level (SRL) and Energy Stress Level (ESL). EA-UAP couples hardware-based PUFs with a modular and layered software architecture, focusing on maintainability and deployability. Formal security verification formally secures against replay, impersonation and man-in-the-middle attacks using Mao–Boyd logic in conjunction with the Scyther tool. The protocol is realized on a Raspberry Pi 4 testbed to verify practical performance. Experimental results demonstrate that EA-UAP can save about 27% of authentication energy costs and 15–20% of end-to-end latency than the baseline UAP, with strong security properties remained. The numerical results indicate that EA-UAP is an efficient and realizable approach for the secure and energy-efficient authentication in 5G UAV networks.