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 to analyze the trade-off between the deployment decisions.
NextGSim[1] is a simulator tool that was developed by our chair to simulate resource management for Beyond 5G Networks. Its main goal is to accurately model resource management algorithms in the radio and computing continuum, to efficiently allocate resources to the users with stringent Quality of Service requirements. The focus of the simulator is latency-sensitive applications, where the end-to-end delay requirement is on the order of tens of milliseconds. These latency-sensitive applications are usually deployed as microservices that form a directed-acyclic graph. The simulator models these microservices according to their processing requirements and the resources they consume.
Accurate modelling of these services is important for scaling and placing decisions. Therefore, it is important to analyze how they behave in different loads and properly model it, so that one can feed it into various algorithms. In this thesis, student is expected to build up on a previous work, where we have developed a monitoring solution for an open source 5G core implementation, free5gc[2] and Aether-on-Ramp[3]. The goal of the thesis is, to extend the measurements using the mentioned setup for various 5G procedures (e.g UE (de)registration, UE configuration update, PDU session establishment) and model the results using statistical techniques.
In the second part of the thesis, student is expected to implement these results into NextGSim, in order to create a simulation environment where users of the simulator can quickly evaluate their algorithms without implementing everything in a real setup.
Prerequisites
- Interest in cellular networks
- Basic knowledge of statistics
- Good knowledge of shell scripting and Python
- Kubernetes knowledge is a plus
References
[1] A. Jano, M. M. Bese, N. Mohan, W. Kellerer and J. Ott, "next GSIM: Towards Simulating Network Resource Management for Beyond 5G Networks," 2023 IEEE Future Networks World Forum (FNWF), Baltimore, MD, USA, 2023, pp. 1-7, doi: 10.1109/FNWF58287.2023.10520638.
[2] free5gc.org
[3] https://docs.aetherproject.org/master/index.html
Contact
Please send your application by email to Mehmet Mert Bese. Please include a grade report and your CV.