Reinforcement Learning Approach for Advanced Sleep Modes Management in 5G Networks

Fatma Ezzahra Salem; Zwi Altman; Azeddine Gati; Tijani Chahed; Eitan Altman

doi:10.1109/VTCFall.2018.8690555

Reinforcement Learning Approach for Advanced Sleep Modes Management in 5G Networks

Fatma Ezzahra Salem
, Zwi Altman
, Azeddine Gati
, Tijani Chahed
, Eitan Altman

CNRS SAMOVAR UMR 5157
Orange Labs
INRIA

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

46 Citations (Scopus)

Abstract

Advanced Sleep Modes (ASMs) correspond to a gradual deactivation of the Base Station (BS)'s components in order to reduce its Energy Consumption (EC). Different levels of Sleep Modes (SMs) can be considered according to the transition time (deactivation and activation durations) of each component. We propose in this paper a management solution for ASMs based on Q-learning approach. The target is to find the optimal durations for each SM level according to the requirements of the network operator in terms of EC reduction and delay constraints. The proposed solution shows that even with a high constraint on the delay, we can achieve high energy savings in a low load scenario (up to 57% of EC reduction) without inducing any impact on the delay. When the delay constraint is relaxed, we can achieve up to almost 90% of energy savings.

Original language	English
Title of host publication	2018 IEEE 88th Vehicular Technology Conference, VTC-Fall 2018 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781538663585
DOIs	https://doi.org/10.1109/VTCFall.2018.8690555
Publication status	Published - 2 Jul 2018
Externally published	Yes
Event	88th IEEE Vehicular Technology Conference, VTC 2018-Fall - Chicago, United States Duration: 27 Aug 2018 → 30 Aug 2018

Publication series

Name	IEEE Vehicular Technology Conference
Volume	2018-August
ISSN (Electronic)	2577-2465

Conference

Conference	88th IEEE Vehicular Technology Conference, VTC 2018-Fall
Country/Territory	United States
City	Chicago
Period	27/08/18 → 30/08/18

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

Keywords

Advanced Sleep Modes
Energy Consumption
Q-learning

Access to Document

10.1109/VTCFall.2018.8690555

Cite this

Salem, F. E., Altman, Z., Gati, A., Chahed, T., & Altman, E. (2018). Reinforcement Learning Approach for Advanced Sleep Modes Management in 5G Networks. In 2018 IEEE 88th Vehicular Technology Conference, VTC-Fall 2018 - Proceedings Article 8690555 (IEEE Vehicular Technology Conference; Vol. 2018-August). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/VTCFall.2018.8690555

@inproceedings{14e8c2a9ba6d4136b2f42dea83e0a23c,

title = "Reinforcement Learning Approach for Advanced Sleep Modes Management in 5G Networks",

abstract = "Advanced Sleep Modes (ASMs) correspond to a gradual deactivation of the Base Station (BS)'s components in order to reduce its Energy Consumption (EC). Different levels of Sleep Modes (SMs) can be considered according to the transition time (deactivation and activation durations) of each component. We propose in this paper a management solution for ASMs based on Q-learning approach. The target is to find the optimal durations for each SM level according to the requirements of the network operator in terms of EC reduction and delay constraints. The proposed solution shows that even with a high constraint on the delay, we can achieve high energy savings in a low load scenario (up to 57\% of EC reduction) without inducing any impact on the delay. When the delay constraint is relaxed, we can achieve up to almost 90\% of energy savings.",

keywords = "Advanced Sleep Modes, Energy Consumption, Q-learning",

author = "Salem, \{Fatma Ezzahra\} and Zwi Altman and Azeddine Gati and Tijani Chahed and Eitan Altman",

note = "Publisher Copyright: {\textcopyright} 2018 IEEE.; 88th IEEE Vehicular Technology Conference, VTC 2018-Fall ; Conference date: 27-08-2018 Through 30-08-2018",

year = "2018",

month = jul,

day = "2",

doi = "10.1109/VTCFall.2018.8690555",

language = "English",

series = "IEEE Vehicular Technology Conference",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2018 IEEE 88th Vehicular Technology Conference, VTC-Fall 2018 - Proceedings",

}

Salem, FE, Altman, Z, Gati, A, Chahed, T & Altman, E 2018, Reinforcement Learning Approach for Advanced Sleep Modes Management in 5G Networks. in 2018 IEEE 88th Vehicular Technology Conference, VTC-Fall 2018 - Proceedings., 8690555, IEEE Vehicular Technology Conference, vol. 2018-August, Institute of Electrical and Electronics Engineers Inc., 88th IEEE Vehicular Technology Conference, VTC 2018-Fall, Chicago, United States, 27/08/18. https://doi.org/10.1109/VTCFall.2018.8690555

Reinforcement Learning Approach for Advanced Sleep Modes Management in 5G Networks. / Salem, Fatma Ezzahra; Altman, Zwi; Gati, Azeddine et al.
2018 IEEE 88th Vehicular Technology Conference, VTC-Fall 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2018. 8690555 (IEEE Vehicular Technology Conference; Vol. 2018-August).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Reinforcement Learning Approach for Advanced Sleep Modes Management in 5G Networks

AU - Salem, Fatma Ezzahra

AU - Altman, Zwi

AU - Gati, Azeddine

AU - Chahed, Tijani

AU - Altman, Eitan

PY - 2018/7/2

Y1 - 2018/7/2

N2 - Advanced Sleep Modes (ASMs) correspond to a gradual deactivation of the Base Station (BS)'s components in order to reduce its Energy Consumption (EC). Different levels of Sleep Modes (SMs) can be considered according to the transition time (deactivation and activation durations) of each component. We propose in this paper a management solution for ASMs based on Q-learning approach. The target is to find the optimal durations for each SM level according to the requirements of the network operator in terms of EC reduction and delay constraints. The proposed solution shows that even with a high constraint on the delay, we can achieve high energy savings in a low load scenario (up to 57% of EC reduction) without inducing any impact on the delay. When the delay constraint is relaxed, we can achieve up to almost 90% of energy savings.

AB - Advanced Sleep Modes (ASMs) correspond to a gradual deactivation of the Base Station (BS)'s components in order to reduce its Energy Consumption (EC). Different levels of Sleep Modes (SMs) can be considered according to the transition time (deactivation and activation durations) of each component. We propose in this paper a management solution for ASMs based on Q-learning approach. The target is to find the optimal durations for each SM level according to the requirements of the network operator in terms of EC reduction and delay constraints. The proposed solution shows that even with a high constraint on the delay, we can achieve high energy savings in a low load scenario (up to 57% of EC reduction) without inducing any impact on the delay. When the delay constraint is relaxed, we can achieve up to almost 90% of energy savings.

KW - Advanced Sleep Modes

KW - Energy Consumption

KW - Q-learning

U2 - 10.1109/VTCFall.2018.8690555

DO - 10.1109/VTCFall.2018.8690555

M3 - Conference contribution

AN - SCOPUS:85064935143

T3 - IEEE Vehicular Technology Conference

BT - 2018 IEEE 88th Vehicular Technology Conference, VTC-Fall 2018 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 88th IEEE Vehicular Technology Conference, VTC 2018-Fall

Y2 - 27 August 2018 through 30 August 2018

ER -

Reinforcement Learning Approach for Advanced Sleep Modes Management in 5G Networks

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Fingerprint

Cite this