Abstract
Airports, being high-risk zones, require robust passenger verification systems. However, current cloud-based facial recognition systems often suffer from single-point-of-failure vulnerabilities. To address this, a blockchain-based airport access mechanism is proposed utilizing Ethereum smart contracts. In this system, passengers requesting access have their facial embeddings verified by a distributed set of Ethereum miners. These miners are organized into clusters, with each cluster using a different facial recognition algorithm, thereby enhancing the accuracy of the verification process. Through asymmetric public- key cryptography and secure hashing, miners reach a consensus both within and across clusters on the authenticity of the passenger. Additionally, the system connects to real-time databases to check for any pending warrants or flying restrictions. Access is granted only if all miners successfully verify and authenticate the passenger. By leveraging a reinforced distributed architecture, this solution eliminates single-point-of-failure risks as well being resistant to DDoS attacks, supporting liveness detection and also accounting for ageing. The experimental Proof-of-Concept demonstrated a 5.72 percent increase in detection accuracy with protocol diversification compared to using a single model.