Quantum-driven security evolution in IoT: AI-powered cryptography and anomaly detection

Quantum-driven security evolution in IoT: AI-powered cryptography and anomaly detection

연구 분야: Cryptography

학회: The Journal of Supercomputing

The rapid expansion of the Internet of Things (IoT) has introduced significant cybersecurity and computational challenges that exceed the capabilities of traditional security mechanisms. This study presents a quantum–classical hybrid security model that enhances anomaly detection, encryption, and predictive maintenance in IoT networks. By integrating quantum key distribution (QKD), post-quantum cryptography (PQC), and quantum machine learning (QML), the proposed approach strengthens threat detection and ensures secure communication. Experimental validation using noisy intermediate-scale quantum (NISQ) devices and IBM quantum simulators demonstrates a 98.7% anomaly detection accuracy, an 80% reduction in latency, and a 3.9% false-positive rate, significantly outperforming traditional AI-based intrusion detection models. The quantum federated learning (QFL) framework further enhances decentralised AI accuracy by 14.5%, while QKD improves encryption resilience by increasing the secure key rate by 500%. The model’s ability to reduce training time by 50% and enhance energy efficiency by 225% makes it scalable for real-time IoT deployments. The proposed security model has wide-ranging implications for industries reliant on IoT networks, such as healthcare, smart cities, and industrial automation, where real-time anomaly detection and secure communication are critical. Organisations deploying IoT infrastructure can leverage quantum-enhanced security to mitigate evolving cyber threats, reduce operational risks, and ensure compliance with future post-quantum cryptographic standards. This research establishes a quantum-secured IoT ecosystem, reinforcing post-quantum encryption and real-time quantum security analytics to mitigate evolving cyber threats. Future directions will explore quantum homomorphic encryption (QHE), zero-trust security architectures (ZTSA), and adaptive quantum AI models, ensuring the practical deployment of quantum-enhanced cybersecurity solutions. The findings highlight quantum methodologies as a scalable, computationally efficient, and high-accuracy approach to securing next-generation IoT environments.