When Digital Twin Meets Generative AI: Intelligent Closed-Loop Network Management

TL;DR

The paper addresses the challenge of intelligent closed-loop network management by integrating generative AI with digital twins to form a GAI-driven Digital Twin (GDT) network architecture. It proposes a three-pronged approach: GDT status emulation, GAI-based feature abstraction, and GAI-based decision-making, coordinated through intelligent external and internal closed loops; it also introduces model-light-weighting, adaptive model selection, and data-model-driven management to tackle overhead and robustness issues. A case study on data-model-driven network management demonstrates QoE gains in multicast video scenarios, highlighting the practical potential of GDT for adaptive resource allocation. The work offers a roadmap for robust, low-overhead, edge-augmented network management and identifies open research directions in module collaboration, specialized models, and resource planning.

Abstract

Generative artificial intelligence (GAI) and digital twin (DT) are advanced data processing and virtualization technologies to revolutionize communication networks. Thanks to the powerful data processing capabilities of GAI, integrating it into DT is a potential approach to construct an intelligent holistic virtualized network for better network management performance. To this end, we propose a GAI-driven DT (GDT) network architecture to enable intelligent closed-loop network management. In the architecture, various GAI models can empower DT status emulation, feature abstraction, and network decision-making. The interaction between GAI-based and model-based data processing can facilitate intelligent external and internal closed-loop network management. To further enhance network management performance, three potential approaches are proposed, i.e., model light-weighting, adaptive model selection, and data-model-driven network management. We present a case study pertaining to data-model-driven network management for the GDT network, followed by some open research issues.

Figures (5)

• Figure 1: The GDT network architecture.
• Figure 2: The specific module interaction in GDT part.
• Figure 3: Procedure of external and internal closed-loop network management.
• Figure 4: Cumulative swipe probability abstracted by the improved autoencoder.
• Figure 5: QoE vs. different computing capacities.