Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

System Failure Prediction within Software 5G Core Networks using Time Series Forecasting

: Chakrabourty, Pousaly; Corici, Marius; Magedanz, Thomas

Postprint urn:nbn:de:0011-n-6383194 (680 KByte PDF)
MD5 Fingerprint: 1862ebd773ea0f86a79d2160481233ef
Erstellt am: 27.7.2021

Institute of Electrical and Electronics Engineers -IEEE-:
IEEE International Conference on Communications Workshops (ICC Workshops) 2021. Proceedings : Virtual Conference, 14 - 23 June 2021
Piscataway, NJ: IEEE, 2021
ISBN: 978-1-7281-9441-7
ISBN: 978-1-7281-9442-4
7 S.
International Conference on Communications (ICC) <2021, Online>
Konferenzbeitrag, Elektronische Publikation
Fraunhofer FOKUS ()
machine learning; time series forecasting; 3GPP 5G Core; Open5GCore toolkit; failure prediction

5G network is very flexible because of the two concepts Network Functions Virtualization (NFV) and the Software Defined Networks (SDN). There are various use cases for 5G technology and for different cases different configuration of the network will be needed. 5G Technology will bring intelligence within the network. The ability to support massive connectivity across diverse devices will result in enormous data volume within the 5G network. Continuous monitoring and traffic log analysis in such a complex architecture will not be sufficient to ensure availability and reliability within the network. The integration of data analytics within the 5G network can leverage the potential of automation. By introducing automation in the monitoring process better Quality of Services (QoS) can be achieved and analysing the network traffic load for better bandwidth utilization within the network. This article proposes a solution to integrate time series based analytics with 5G core and predicting any threats within the system which can lead to system failure. To validate the proposal Fraunhofer FOKUS Open5GCore toolkit is used.