Skip to Main content Skip to Navigation
New interface
Theses

4G/5G cellular networks metrology and management

Imane Oussakel 1 
1 LAAS-SARA - Équipe Services et Architectures pour Réseaux Avancés
LAAS - Laboratoire d'analyse et d'architecture des systèmes
Abstract : The proliferation of sophisticated applications and services comes with diverse performance requirements as well as an exponential traffic growth for both upload and download. The cellular networks such as 4G and 5G are advocated to support this diverse and huge amount of data. This thesis work targets the enforcement of advanced cellular network supervision and management techniques taking the traffic explosion and diversity as two main challenges in these networks. The first contribution tackles the intelligence integration in cellular networks through the estimation of users uplink instantaneous throughput at small time granularities. A real time 4G testbed is deployed for such aim with an exhaustive metrics benchmark. Accurate estimations are achieved.The second contribution enforces the real time 5G slicing from radio resources perspective in a multi-cell system. For that, two exact optimization models are proposed. Due to their high convergence time, heuristics are developed and evaluated with the optimal models. Results are promising, as two heuristics are highly enforcing the real time RAN slicing.
Complete list of metadata

https://theses.hal.science/tel-03236078
Contributor : ABES STAR :  Contact
Submitted on : Wednesday, May 26, 2021 - 10:20:08 AM
Last modification on : Tuesday, October 25, 2022 - 11:58:11 AM
Long-term archiving on: : Friday, August 27, 2021 - 6:36:04 PM

File

2020TOU30261b.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-03236078, version 1

Citation

Imane Oussakel. 4G/5G cellular networks metrology and management. Networking and Internet Architecture [cs.NI]. Université Paul Sabatier - Toulouse III, 2020. English. ⟨NNT : 2020TOU30261⟩. ⟨tel-03236078⟩

Share

Metrics

Record views

127

Files downloads

96