A Security-Aware Multi-Objective Optimization Framework for AI-Driven SDN Networks: Comparative Analysis of Constrained Solvers
Dijlah Journal of Engineering Sciences ISSN: 3078-9664, e-ISSN: 3078-9656 · 17.06.2026 · 3 (2)
DOI: 10.65204/djes.v3i2.665 ↗Kurz gefasst
Fachartikel in der Zeitschrift Dijlah Journal of Engineering Sciences ISSN: 3078-9664, e-ISSN: 3078-9656, online erschienen am 17.06.2026. Verfasst von Susan AlKarawi und 1 weiteren Autor:innen.
Abstract (Originalsprache)
The fast development of modern networks have raised the importance for flexible, smart and security-conscious management. Software-Defined Networking (SDN) and Artificial Intelligence (AI) are new possibilities for dynamically optimised, secure network resource provisioning to combat these challenges. In this paper, we put forward a security-aware multi-objective optimization model for AI-powered SDN networks that co-optimizes important performance metrics including end-to-end delay, packet loss, throughput as well as the security risk. The model reflects realistic constraints like flow conservation, link capacity, Quality of Service (QoS), and security-aware service-oriented SLAs. Four different optimization methods have been applied to solve the resulting nonlinear constrained optimization problem: classical Gradient Descent (GD), Sequential Quadratic Programming (SQP), Active-Set, and Interior-Point methods. It is shown through simulation results that convergence of GD is slow and leads to large constraint violations while the proposed constrained solvers enjoy much better performance. Specifically, the Interior-Point method reveals faster convergence and fewer constraint violate characteristics, rendering it suitable for large-scale and security-sensitive SDN application. The findings highlight the significance of constraint-aware optimization in secure and high-performance AI powered networks.
Zitieren
APA
AlKarawi, S., & Mohammed, R. (2026). A Security-Aware Multi-Objective Optimization Framework for AI-Driven SDN Networks: Comparative Analysis of Constrained Solvers. Dijlah Journal of Engineering Sciences ISSN: 3078-9664, e-ISSN: 3078-9656, 3(2). https://doi.org/10.65204/djes.v3i2.665
Vancouver
AlKarawi S, Mohammed R. A Security-Aware Multi-Objective Optimization Framework for AI-Driven SDN Networks: Comparative Analysis of Constrained Solvers. Dijlah Journal of Engineering Sciences ISSN: 3078-9664, e-ISSN: 3078-9656. 2026;3(2). doi:10.65204/djes.v3i2.665
BibTeX
@article{AlKarawi2026,
author = {AlKarawi, Susan and Mohammed, Rasool},
title = {A Security-Aware Multi-Objective Optimization Framework for AI-Driven SDN Networks: Comparative Analysis of Constrained Solvers},
journal = {Dijlah Journal of Engineering Sciences ISSN: 3078-9664, e-ISSN: 3078-9656},
year = {2026},
volume = {3},
number = {2},
doi = {10.65204/djes.v3i2.665}
}
EUVIMED