The autonomous systems field, including UAVs, HAPs, and satellites, has rapidly expanded in recent years. As such, 3D visualization in a 3D environment is a promising method for visualizing networking in autonomous systems to improve management and monitoring.

Our research proposes a digital twin solution for Munich that uses 3D visualization, network virtualization, and machine learning algorithms to simulate and optimize the behavior of the city's network functions and radio functions. We begin by describing the fundamentals of digital twin technology and model a resource orchestration problem to optimize the network performance of non-standalone (NSA) 5G from user equipment to the core network. We then examine the specific challenges faced by Munich in managing its network functions and radio functions. Our solution utilizes 3D visualization to provide an immersive and comprehensive view of the city's network functions and radio functions, allowing operators to quickly identify and address issues. Optimization and machine learning algorithms analyze the network data to optimize performance and provide real-time recommendations to prevent failures and operate the networks. We validate our approach's performance and discuss its potential impact on Munich, including improved network performance, better resource management, and increased efficiency and safety. We also highlight its potential application in disaster situations, inspired by the data based on Munich floods of July 2021. Overall, we argue that using a digital twin approach to virtualize network functions and radio functions within a 3D simulation has significant potential to enhance network management and optimization. The Munich case study demonstrates the feasibility and potential benefits of this approach for other cities facing similar challenges.

Menu

Dr. CHUAN PHAM