Learning a Network Digital Twin as a Hybrid System

2025-10-31

AI Summary

To address the challenge of accurately modeling communication quality under user mobility in multi-cell dynamic wireless networks, this paper proposes a hybrid system-driven Network Digital Twin (NDT) framework tailored for 6G. The framework structurally abstracts time-varying topologies and mobility scenarios by decomposing the network into a base-station-associated primary mode and spatial sub-modes that capture similar channel characteristics. It integrates real-time user measurement data and employs an annealing-optimized learning algorithm to enable online model identification, incremental updates, and lightweight deployment. Experimental evaluation on a real-world dual-cell 5G testbed demonstrates that the proposed approach significantly reduces memory footprint, computational overhead, and data requirements, while simultaneously improving modeling accuracy and dynamic adaptability. The results validate both its effectiveness and real-time performance advantages.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Network digital twin (NDT) models are virtual models that replicate the behavior of physical communication networks and are considered a key technology component to enable novel features and capabilities in future 6G networks. In this work, we focus on NDTs that model the communication quality properties of a multi-cell, dynamically changing wireless network over a workspace populated with multiple moving users. We propose an NDT modeled as a hybrid system, where each mode corresponds to a different base station and comprises sub-modes that correspond to areas of the workspace with similar network characteristics. The proposed hybrid NDT is identified and continuously improved through an annealing optimization-based learning algorithm, driven by online data measurements collected by the users. The advantages of the proposed hybrid NDT are studied with respect to memory and computational efficiency, data consumption, and the ability to timely adapt to network changes. Finally, we validate the proposed methodology on real experimental data collected from a two-cell 5G testbed.

Problem

• Modeling wireless network communication quality dynamically
• Learning hybrid digital twin through optimization and data
• Validating memory-efficient network adaptation using real data

Innovation

• Hybrid system modeling network with base station modes
• Annealing optimization learning from online user data
• Memory efficient adaptation to dynamic network changes