The paper presents a comprehensive approach to calibrating and automating a 5G system-level simulator, Simu5G, to enable faster innovation in the 5G domain.
The key highlights are:
Calibration of Simu5G: The authors calibrate Simu5G, an open-source 5G simulator, following the 3GPP guidelines and standards for both urban and rural deployment scenarios. This ensures the simulator provides a realistic and dependable evaluation platform.
Automation of Simu5G configuration: The authors develop a YAML-based API to automatically generate Simu5G configurations, reducing the steep learning curve and configuration complexity associated with the simulator. Users only need to provide high-level topological information, and the tool handles the underlying architectural details.
Demonstration of use case: The authors showcase the usability of the calibrated and automated Simu5G by developing a neural network-based anomaly detection model in a 5G Radio Access Network (RAN). The model is evaluated using the data generated from the calibrated simulator.
Open-source release: The authors share the developed solutions, including the calibrated simulator and the automation tool, on a public Git repository for reproducibility and extension.
Overall, the paper presents a holistic approach to calibrating and automating a 5G simulator, which can significantly accelerate innovation in the 5G domain by providing a realistic evaluation platform and reducing the barriers to simulator usage.
To Another Language
from source content
arxiv.org
Key Insights Distilled From
by Conrado Boei... at arxiv.org 04-17-2024
https://arxiv.org/pdf/2404.10643.pdfDeeper Inquiries