AI-Enhanced Cybersecurity for IoT Environments

Authors

  • Ayesha Fatima Department of Computer Science, National University of Sciences & Technology (NUST), Islamabad, Pakistan
  • Hassan Raza Centre for Cybersecurity, University of the Punjab, Lahore, Pakistan
  • Sana Javed Department of Electrical & Computer Engineering, NED University of Engineering & Technology, Karachi, Pakistan

Keywords:

IoT security, anomaly detection, federated learning, zero-trust, MLOps, intrusion detection, device identity, threat intelligence

Abstract

The proliferation of Internet-of-Things (IoT) devices across critical domains—healthcare, energy, transportation, and manufacturing—has expanded the cyber-attack surface beyond traditional enterprise perimeters. This article synthesizes AI-driven mechanisms that harden IoT ecosystems, including anomaly-based intrusion detection, federated learning for privacy-preserving analytics, automated firmware integrity checks, and graph-based threat hunting across edge–cloud topologies. We present a layered reference architecture covering device, edge, and cloud planes with zero-trust policies and security digital twins for continuous verification. We also discuss governance (ETSI EN 303 645, NISTIR 8259, IEC 62443), data stewardship, and MLOps for security models. The paper concludes with strategic guidance for Pakistani industry adoption and a benchmark-style illustrative figure showing relative risk reduction by AI techniques.

Downloads

Published

2022-12-31

Issue

Jul-Dec 2022

Section

Articles

License

All articles published in the International Journal of Advanced Sciences and Computing (IJASC) are open access and licensed under the Creative Commons Attribution (CC BY 4.0) license. Authors retain copyright, and users may read, share, reuse, and adapt the content with proper attribution.