TechRxiv
An SDN NFV Proof-of-Concept Test-bed for ML-based Ntework Management.pdf (997.6 kB)
Download file

An SDN/NFV Proof-of-Concept Test-Bed for Machine Learning-based Network Management

Download (997.6 kB)
preprint
posted on 04.05.2021, 13:59 by Wei JiangWei Jiang, Mathias Strufe, Michael Gundall, Hans Dieter Schotten
Complexity and heterogeneity of the fifth generation (5G) and beyond mobile systems impose a great challenge on current network managing approaches, which are vulnerable, time-consuming and costly. The state-of-the-art research direction in this field is to apply machine learning (ML) techniques to realize intelligent and highly self-organized networking. Unlike the physical layer, theoretical analyses and numerical simulations on the management layer are generally infeasible or not scientifically rigorous enough. Therefore, in this paper, we present a software-defined and virtualized wireless test-bed that is established to evaluate ML-based network management. Based on open-source software and off-the-shelf hardware, this test-bed is easily reproducible, which in turn is hopeful to foster innovative works in this field.

Funding

Tacnet4.0

History

Email Address of Submitting Author

wei.jiang@dfki.de

ORCID of Submitting Author

https://orcid.org/0000-0002-3719-3710

Submitting Author's Institution

German Research Center for Artificial Intelligence (DFKI)

Submitting Author's Country

Germany