Building a Network Data Analytics Function for 5G Using Open Telemetry

Topic Overview

Cloud-native 5G is an emerging topic that is enabled by the new architecture of 5G, cloud computing and virtualization. New service-based architecture of cellular networks allows core network components to be deployed as microservices , conformant to cloud-native principles. This approach paves the way for new research to efficiently automate, scale, and verify 5G core deployments.

Open Telemetry is an observability tool that is built for the cloud. It has extensive features in order to measure various metrics such as CPU usage, request duration etc. It paves up new ways to observe how your services are performing and take actions proactively on them. And since the cloud native 5G comprises of microservices, same principles go for it.

The goal of this thesis is to build a Network Data Analytics Function(NWDAF)[1] on an existing 5G testbed for network function scaling problems. The student is expected to build a NWDAF on an existing 5G core testbed in our chair, powered by Aether[2], free5GC[3] and OpenTelemetry[4] and generate predictions about the network performance using predictive AI models.

Even though, NWDAF can be used for many purposes, it's not feasible to implement all of them in the scope of a master thesis. Therefore, student is expected to build a proactive horizontal pod autoscaler(HPA) using NWDAF as a use case, while keeping the implementation generic such that new features can be added. The thesis will be concluded by comparing NWDAF's performance to the built-in reactive HPA of Kubernetes.

Prerequisites
- Interest in cellular networks
- Kubernetes and Open Telemetry experience is a plus
- AI model deployment experience is a plus

References
[1] free5gc.org/blog/20241127/20241127/

[2] docs.aetherproject.org/master/index.html

[3] free5gc.org

[4] opentelemetry.io

Contact 

Please send your application by email to Mehmet Mert Bese. Please include a grade report and your CV.