In this paper we evaluate the performance impact of introducing the new deep sleep state features that may be deployed in 5G NR base stations. We evaluate the effects on traffic performance and energy efficiency by applying an updated power consumption model. Previously proposed power models may be too optimistic and thus the impact of a new more conservative power model has been investigated. A sensitivity analysis is also performed to understand the effects on the KPIs with various parameter settings of the power model. In the evaluations we show that the energy savings that is achievable by using a more conservative power model is still very high and that the impact on user performance can be small if the sleep states are applied in a thoughtful manner. Up to 80% energy savings is possible to achieve in a 5G hetnet scenario. The performance impact can be limited to a few percent extra delay on file transmissions. It is also found that even quite large modifications to the power model give similar results.
KPI Impact on 5G NR Deep Sleep State Adaption
2019-09-01
1145867 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Input Image Adaption for Robust Direct SLAM using Deep Learning
Deutsches Zentrum für Luft- und Raumfahrt (DLR) | 2020
|ROSE: Robust State Estimation via Online Covariance Adaption
Springer Verlag | 2023
|Online Contents | 2009
METHOD FOR RESTRAINT DEPLOYMENT ADAPTION AND SYSTEM FOR RESTRAINT DEPLOYMENT ADAPTION
Europäisches Patentamt | 2024
|