Enhancing service classification for network slicing in 5G using machine learning algorithms

Mohammedali, Noor Abdalkarem, Kanakis, Triantafyllos, Al-Sherbaz, Ali ORCID: 0000-0002-0995-1262, Agyeman, Michael Opoku and Hasson, Saad Talib (2023) Enhancing service classification for network slicing in 5G using machine learning algorithms. In: 6th International Conference, New Trends in Information and Communications Technology Applications, 16-17 November 2022, Baghdad, Iraq. ISSN 1865-0929 ISBN 9783031354410

Text (Peer-reviewed version) 12955 AL-SHERBAZ Ali (2023) Enhancing Service Classification for Network Slicing article.pdf - Accepted Version Restricted to Repository staff only until 10 June 2024. (Publisher Embargo). Available under License All Rights Reserved. Download (1MB)

Abstract

In a virtualization aspect, Network Function Virtualization (NFV) has a role in implementing network slicing. Using NFV to slice the network, make the network more flexible, but very complicated in term of management. A slice is a set of services that the network needs based on the user requirements. Moreover, each slice has a set of services called sub-slice, or one type of service. This research aims to improve the availability and scalability of the services in network slicing by managing the performance of the inter/intra slice in real-time. Also, this research will enhance the Quality of Service (QoS) for the network resources and services and the Quality of Experience (QoE) for the users within the slice when we applied machine learning algorithms to classify and predicate accurate service to the user. With this research, we implemented the slices based on the principles of NFV to deliver flexibility in the 5G network by creating multiple slices on top of the physical network. When the implementation of the prototype is completed, traffic generated tool was used to send traffic over the slices. After data collection, we classified different services using machine learning algorithms. The optimizable tree model had almost high accuracy among other algorithms which was 99.3%

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	5G; Network slicing; NSSF; Traffic-Classification; QoS; Inter-slice; NFV; E2E; Machine learning
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software
Divisions:	Schools and Research Institutes > School of Business, Computing and Social Sciences
Research Priority Areas:	Applied Business & Technology
Depositing User:	Rhiannon Goodland
Date Deposited:	20 Jul 2023 08:24
Last Modified:	24 Apr 2024 09:45
URI:	https://eprints.glos.ac.uk/id/eprint/12955

University Staff: Request a correction | Repository Editors: Update this record