Secure and decentralised swarm authentication using hardware security primitives

Abstract

Autonomous drone swarms are increasingly deployed in critical domains such as infrastructure inspection, environmental monitoring, and emergency response. While their distributed operation enables scalability and resilience, it also introduces new vulnerabilities, particularly in authentication and trust establishment. Conventional cryptographic solutions, including public key infrastructures (PKI) and symmetric key protocols, impose computational and connectivity requirements unsuited to resource-constrained and external infrastructure-free swarm deployments. In this paper, we present a decentralized authentication scheme rooted in hardware security primitives (HSPs); specifically, Physical Unclonable Functions (PUFs) and True Random Number Generators (TRNGs). The protocol leverages master-initiated token broadcasting, iterative HSP seed evolution, randomized response delays, and statistical trust evaluation to detect cloning, replay, and impersonation attacks without reliance on centralized authorities or pre-distributed keys. Simulation studies demonstrate that the scheme achieves lightweight operation, rapid anomaly detection, and robustness against wireless interference, making it well-suited for real-time swarm systems.