EUVIMEDEUVIMED entdecken
Crossref

Intelligent Multi-Path Cooperative Routing for Resource-Constrained Software-Defined Networks

Jiawei Wu, ZeLin Zhu, Yibo Wang, Hongzhang Si

IEEE Transactions on Cognitive Communications and Networking · 2026 · 12 · 8123-8140

DOI: 10.1109/tccn.2026.3692140 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Cognitive Communications and Networking, 2026. Verfasst von Jiawei Wu und 3 weiteren Autor:innen. Die Arbeit nennt 41 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Intelligent multi-path routing in resource-constrained Software-Defined Networks (SDNs) must balance real-time adaptability with low operational overhead. Existing centralized or static methods often incur high latency, training costs, and inter-controller conflicts. To address these challenges, this paper proposes CACAR (Classification-Augmented Cooperative Adaptive Routing), a novel framework that integrates traffic classification with topology-aware cooperative learning. CACAR employs offline clustering to pre-compute routing strategies from historical traffic patterns and uses a lightweight online classifier to match live flows to optimal strategies, minimizing online computation. For dynamic traffic, a Graph Neural Network-enhanced multi-agent reinforcement learning scheme cooperatively adjusts path weights across distributed controllers. A resource-aware asymmetric Nash bargaining mechanism is introduced to resolve inter-controller conflicts equitably. Extensive simulations demonstrate that CACAR significantly reduces route update latency and inter-controller conflict rates by over 60% compared to DRL and ECMP baselines. Furthermore, it substantially reduces the volume of online DRL computation and controller processing compared to the centralized DRL baseline. All these improvements are achieved while maintaining robust performance across diverse scenarios with low overhead.

Zitieren

APA

Wu, J., Zhu, Z., Wang, Y., & Si, H. (2026). Intelligent Multi-Path Cooperative Routing for Resource-Constrained Software-Defined Networks. IEEE Transactions on Cognitive Communications and Networking, 12, 8123-8140. https://doi.org/10.1109/tccn.2026.3692140

Vancouver

Wu J, Zhu Z, Wang Y, Si H. Intelligent Multi-Path Cooperative Routing for Resource-Constrained Software-Defined Networks. IEEE Transactions on Cognitive Communications and Networking. 2026;12:8123-8140. doi:10.1109/tccn.2026.3692140

BibTeX
@article{Wu2026,
  author = {Wu, Jiawei and Zhu, ZeLin and Wang, Yibo and Si, Hongzhang},
  title = {Intelligent Multi-Path Cooperative Routing for Resource-Constrained Software-Defined Networks},
  journal = {IEEE Transactions on Cognitive Communications and Networking},
  year = {2026},
  volume = {12},
  pages = {8123--8140},
  doi = {10.1109/tccn.2026.3692140}
}
Kostenlos starten →