Hierarchical Network Digital Twin: Abstraction is All You Need

Abstract

The wireless industry has traditionally pursued higher data rates as a primary objective. However, the focus has shifted from merely increasing throughput to enabling a wide range of emerging use cases and service types, each with distinct quality-of-service (QoS) requirements. Managing such heterogeneous and dynamic networks is becoming increasingly challenging due to their growing complexity, the need for faster control loops, and the diversity of service demands. Network Digital Twins (NDTs) have emerged as a promising paradigm to address these challenges by providing virtual replicas of real networks for observation, prediction, and optimization.

Yet, existing NDT approaches often struggle to simultaneously capture multiple facets of network behavior that inherently require different levels of data granularity, precision, and scale. To overcome this limitation, we propose a Hierarchical Network Digital Twin (H-NDT) architecture. By leveraging appropriate levels of abstraction, the H-NDT captures essential network characteristics while preserving scalability and computational efficiency. Furthermore, we discuss the interaction and synchronization mechanisms between different hierarchical levels, enabling coherent cross-layer network understanding and control.

Short Bio

Ehsan Tohidi (Member, IEEE) received the B.Sc., M.Sc., and Ph.D. degrees (Hons.) in electrical engineering (communications and signal processing) from Sharif University of Technology, Tehran, Iran, in 2011, 2013, and 2018, respectively. He is currently the Head of the Sustainable and Modular Networks Group at the Fraunhofer Institute for Telecommunications, Heinrich Hertz Institute (HHI), Berlin, Germany, and a Senior Researcher and Project Coordinator at the Technical University of Berlin, Germany. From 2019 to 2020, he was a Postdoctoral Researcher at EURECOM, Biot, France, and from September 2016 to March 2017, he was a Visiting Research Associate at Delft University of Technology, Delft, The Netherlands.

He and his group are actively involved in the German 6G Research and Innovation Cluster (6G-RIC), coordinated by HHI, as well as several related national and international research projects. He leads and coordinates multiple academic and industry collaborations and has organized and chaired workshops and sessions at major conferences, including the Berlin 6G Conference and IEEE GLOBECOM. He has authored more than 30 peer-reviewed publications, including high-impact IEEE Transactions papers and top-tier IEEE conference papers. His research interests include O-RAN, digital twins, radar and sensing, signal processing, discrete and continuous optimization, and machine learning.

Venue