Willame, Martin
[UCL]
Oestges, Claude
[UCL]
Vandendorpe, Luc
[UCL]
With the growing demand in high data rate and mobile subscriptions triggered by the development of the 5th generation of mobile networks, the power consumption of cellular networks is expected to increase. A growing interest towards greener wireless telecommunications has therefore been observed, with the view to cut down the carbon footprint of the whole sector. To this end, a potential solution is to introduce sleep modes strategies for the base stations. In this thesis, the benefit in terms of energy efficiency of activity based sleeping strategies is studied using stochastic geometry. A characterisation of the load, defining the activity of each base stations, is performed for different network configurations involving homogeneous and/or clustered distribution of users. Different features present in 5G network such as inhomogeneous networks, beamforming and millimeter wave transmission have been taken into account in the mathematical expressions developed herein to characterise the energy efficiency. The present work gives an overview of the impact of various network parameters on the energy efficiency and shows the benefit of load-dependent sleeping strategies. To sum up, this thesis provides a general framework to better understand the characteristics of greener 5G networks.
Bibliographic reference |
Willame, Martin. Stochastic geometry-based sleep mode strategies for energy efficient cellular networks. Ecole polytechnique de Louvain, Université catholique de Louvain, 2021. Prom. : Oestges, Claude ; Vandendorpe, Luc. |
Permanent URL |
http://hdl.handle.net/2078.1/thesis:33173 |