CBL - Campus del Baix Llobregat

Projecte llegit

Títol: Design and analysis of fully virtualized cellular networks based on open-source frameworks


Departament: ENTEL

Títol: Design and analysis of fully virtualized cellular networks based on open-source frameworks

Data inici oferta: 19-01-2021     Data finalització oferta: 19-09-2021

Estudis d'assignació del projecte:
    MU MASTEAM 2015
Tipus: Individual
Lloc de realització: Fora UPC    
        Supervisor extern: Dr. Estefanía Coronado Calero
        Institució/Empresa: Fundació I2CAT
        Titulació del Director: PhD in Advanced Computer Technologies
Nom del segon director (UPC): Shuaib Siddiqui (i2Cat)
Departament 2n director:
Paraules clau:
Cellular networks, Virtualization Frameworks, Kubernetes , srsLTE, Open5GS
Descripció del contingut i pla d'activitats:
The intern will become a member of the Software Networks group at i2CAT. SN currently takes part in several H2020 projects developing 5G technologies, and the work of the intern would be related to one of these projects called AI@EDGE.

The project will focus on the design and implementation of an automated profiling tool to be embedded on a MEC platform to quantify the performance and resources needed by the AI functions of different nature (e.g., ML, DL, online learning, etc.) running at the networks' edge.

The work of the intern will be structured in the following phases:

1) Get familiar with the ETSI MEC architecture and the current design of the MEC Platform,

2) Get familiar with profiling tools in the case of not having previous experience,

3) Design and implement at the MEC platform a profiling tool that admit different AI pipelines (e.g., ML, DL, online learning, etc.),

4) Evaluate the performance of the system, and

5) Compile the results in an academic publication. The intern will be supported by the various team members in this mission.
Overview (resum en anglès):
Virtualization of cellular networks is one of the key areas of research where technologies, infrastructure and challenges are rapidly changing as 5G system architecture demands a paradigm shift. Adding these necessities to the appearance of new technologies such as Software Defined Networks (SDN) and Network Function Virtualization (NFV), encourage the seeking for the implementation of a more flexible and adaptive architecture, for mobile cellular networks.
In this Master Thesis, a comprehensive view is provided upon various scenarios to enable the deployment of a fully distributed and open-source cellular network, as well as to provide an analysis on the impact that both Radio Access Network (RAN) and Core Network have on the resource utilization (radio and virtualized infrastructure) as the network conditions vary. The prototype proposed has been performed using a 4G setup given the currently limited availability of 5G setup and its high cost and compare with a state-of-the-art deployment using a different virtualization tool. Nevertheless, the results presented in this document could be extended to the 5G scenarios. The reason for this is that this Master Thesis aims to study the viability and the performance of virtualized infrastructure to host the aforementioned network segments, and the frameworks used implement both the 4G and the 5G stacks, and even share the same software modules to implement the network functions regardless of the cellular network.
The analysis of the gathered data expose that fully virtualized cellular network deployments present better performance in case of the flexibility, low setting time and ease to deploy, while keeping the same level of resources usage as the non-virtualized deployments. Furthermore, some future works and directions for research to and develop more flexible and more adaptive deployments are proposed, as well as to expand the analysis with higher network capacities available in 5G.

© CBLTIC Campus del Baix Llobregat - UPC