Aegis-Edge: Gated Agentic Cyber Defense with Deception-Oriented Control for IoT/IIoT Edge Systems
International Journal of Creative and Open Research in Engineering and Management · 08.05.2026 · 02 (05) · 1-9
DOI: 10.55041/ijcope.v2i5.262 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Creative and Open Research in Engineering and Management, online erschienen am 08.05.2026. Verfasst von Anjali Srivastava Anjali Srivastava.
Abstract (Originalsprache)
Agentic AI is shifting cybersecurity from passive, reactive classification toward systems that reason, plan, and act across extended operational loops. Yet existing defenses remain constrained by high computational cost, excessive response la-tency, and unresolved questions of safe autonomy, particularly in IoT and Industrial IoT (IIoT) edge environments. This paper proposes Aegis-Edge, a novel agentic cyber-defense frame-work that integrates lightweight anomaly sensing, uncertainty-triggered reasoning, and deception-aware response orchestration specifically designed for resource-constrained edge deployments. The framework is motivated by a clear gap in the literature: while AI-for-cybersecurity surveys outline the need for advanced methods and new infrastructures, and human-in-the-loop XAI studies emphasize transparent decision support, neither provides a closed-loop autonomous architecture for the edge. Aegis-Edge addresses this gap through a gated four-agent pipeline: a compact Monitor Agent for always-on anomaly scoring; a Reasoner Agent activated only under uncertainty, concept drift, or escalating attack severity; a Deception Agent that selects bounded response primitives including canary tokens, honeypot redirec-tion, and decoy credential injection; and a Governor Agent that enforces policy constraints and human-approval thresholds. We describe the system architecture, formal optimization objective, experimental protocol, and an illustrative simulation study over standard public intrusion datasets. Simulated results indicate that the gated design reduces end-to-end response latency and energy consumption substantially relative to monolithic agentic baselines, while maintaining recall above 97.8% and reducing false positive rate to 1.1%. These findings position Aegis-Edge as a credible foundation for operationally viable, safe-by-design autonomous cyber defense at the edge. Index Terms—agentic AI, cyber deception, edge security, in-trusion detection, IoT/IIoT, resource-constrained systems, multi-agent systems, autonomous response
Zitieren
APA
Srivastava, A. S. A. (2026). Aegis-Edge: Gated Agentic Cyber Defense with Deception-Oriented Control for IoT/IIoT Edge Systems. International Journal of Creative and Open Research in Engineering and Management, 02(05), 1-9. https://doi.org/10.55041/ijcope.v2i5.262
Vancouver
Srivastava ASA. Aegis-Edge: Gated Agentic Cyber Defense with Deception-Oriented Control for IoT/IIoT Edge Systems. International Journal of Creative and Open Research in Engineering and Management. 2026;02(05):1-9. doi:10.55041/ijcope.v2i5.262
BibTeX
@article{Srivastava2026,
author = {Srivastava, Anjali Srivastava Anjali},
title = {Aegis-Edge: Gated Agentic Cyber Defense with Deception-Oriented Control for IoT/IIoT Edge Systems},
journal = {International Journal of Creative and Open Research in Engineering and Management},
year = {2026},
volume = {02},
number = {05},
pages = {1--9},
doi = {10.55041/ijcope.v2i5.262}
}
EUVIMED