Combined learning for energy efficiency in heterogeneous cellular networks

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

1 Citation (Scopus)

Abstract

In this paper, we investigate improving energy ef?ciency in heterogeneous cellular networks (HCNs). A Stackelberg learning game is ?rst formulated, in which the macrocells behave as the leaders and the small-cells are followers. In the beginning of each epoch (every T time slots are de?ned as one epoch), the leaders update their power adaptation policies by knowing the best-responses of all followers, while the followers compete against each other in each time slot with only the leaders' action information. The hierarchy in learning procedure indicates the macrocell states in any two consecutive epochs are highly correlated. Then the small-cells' historical policy information can be leveraged to enhance the learning performance. Accordingly, a combined learning framework is established, through combining the Stackelberg learning formulation and the technique of transfer learning, to tell players how to plan the action decisions. Simulations presented show that the combined learning algorithm substantially improves the energy ef?ciency of HCNs.
Original languageEnglish
Title of host publicationGreen Cellular 2013
Subtitle of host publication24th International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC Workshops 2013
PublisherInstitute of Electrical and Electronic Engineers IEEE
Pages21-25
ISBN (Electronic)978-1-4799-0122-7
DOIs
Publication statusPublished - 2013
MoE publication typeNot Eligible
Event24th Annual IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC'13 - London, United Kingdom
Duration: 8 Nov 201311 Nov 2013
Conference number: 24

Conference

Conference24th Annual IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC'13
Abbreviated titlePIMRC'13
CountryUnited Kingdom
CityLondon
Period8/11/1311/11/13

Fingerprint

Learning algorithms
Energy efficiency

Cite this

Chen, X., Zhang, H., & Lasanen, M. (2013). Combined learning for energy efficiency in heterogeneous cellular networks. In Green Cellular 2013: 24th International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC Workshops 2013 (pp. 21-25). Institute of Electrical and Electronic Engineers IEEE. https://doi.org/10.1109/PIMRCW.2013.6707829
Chen, Xianfu ; Zhang, H. ; Lasanen, Mika. / Combined learning for energy efficiency in heterogeneous cellular networks. Green Cellular 2013: 24th International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC Workshops 2013. Institute of Electrical and Electronic Engineers IEEE, 2013. pp. 21-25
@inproceedings{31e7e6cbb2ec489b8c7c693623c11762,
title = "Combined learning for energy efficiency in heterogeneous cellular networks",
abstract = "In this paper, we investigate improving energy ef?ciency in heterogeneous cellular networks (HCNs). A Stackelberg learning game is ?rst formulated, in which the macrocells behave as the leaders and the small-cells are followers. In the beginning of each epoch (every T time slots are de?ned as one epoch), the leaders update their power adaptation policies by knowing the best-responses of all followers, while the followers compete against each other in each time slot with only the leaders' action information. The hierarchy in learning procedure indicates the macrocell states in any two consecutive epochs are highly correlated. Then the small-cells' historical policy information can be leveraged to enhance the learning performance. Accordingly, a combined learning framework is established, through combining the Stackelberg learning formulation and the technique of transfer learning, to tell players how to plan the action decisions. Simulations presented show that the combined learning algorithm substantially improves the energy ef?ciency of HCNs.",
author = "Xianfu Chen and H. Zhang and Mika Lasanen",
note = "Project code: 82164",
year = "2013",
doi = "10.1109/PIMRCW.2013.6707829",
language = "English",
pages = "21--25",
booktitle = "Green Cellular 2013",
publisher = "Institute of Electrical and Electronic Engineers IEEE",
address = "United States",

}

Chen, X, Zhang, H & Lasanen, M 2013, Combined learning for energy efficiency in heterogeneous cellular networks. in Green Cellular 2013: 24th International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC Workshops 2013. Institute of Electrical and Electronic Engineers IEEE, pp. 21-25, 24th Annual IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC'13, London, United Kingdom, 8/11/13. https://doi.org/10.1109/PIMRCW.2013.6707829

Combined learning for energy efficiency in heterogeneous cellular networks. / Chen, Xianfu; Zhang, H.; Lasanen, Mika.

Green Cellular 2013: 24th International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC Workshops 2013. Institute of Electrical and Electronic Engineers IEEE, 2013. p. 21-25.

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

TY - GEN

T1 - Combined learning for energy efficiency in heterogeneous cellular networks

AU - Chen, Xianfu

AU - Zhang, H.

AU - Lasanen, Mika

N1 - Project code: 82164

PY - 2013

Y1 - 2013

N2 - In this paper, we investigate improving energy ef?ciency in heterogeneous cellular networks (HCNs). A Stackelberg learning game is ?rst formulated, in which the macrocells behave as the leaders and the small-cells are followers. In the beginning of each epoch (every T time slots are de?ned as one epoch), the leaders update their power adaptation policies by knowing the best-responses of all followers, while the followers compete against each other in each time slot with only the leaders' action information. The hierarchy in learning procedure indicates the macrocell states in any two consecutive epochs are highly correlated. Then the small-cells' historical policy information can be leveraged to enhance the learning performance. Accordingly, a combined learning framework is established, through combining the Stackelberg learning formulation and the technique of transfer learning, to tell players how to plan the action decisions. Simulations presented show that the combined learning algorithm substantially improves the energy ef?ciency of HCNs.

AB - In this paper, we investigate improving energy ef?ciency in heterogeneous cellular networks (HCNs). A Stackelberg learning game is ?rst formulated, in which the macrocells behave as the leaders and the small-cells are followers. In the beginning of each epoch (every T time slots are de?ned as one epoch), the leaders update their power adaptation policies by knowing the best-responses of all followers, while the followers compete against each other in each time slot with only the leaders' action information. The hierarchy in learning procedure indicates the macrocell states in any two consecutive epochs are highly correlated. Then the small-cells' historical policy information can be leveraged to enhance the learning performance. Accordingly, a combined learning framework is established, through combining the Stackelberg learning formulation and the technique of transfer learning, to tell players how to plan the action decisions. Simulations presented show that the combined learning algorithm substantially improves the energy ef?ciency of HCNs.

U2 - 10.1109/PIMRCW.2013.6707829

DO - 10.1109/PIMRCW.2013.6707829

M3 - Conference article in proceedings

SP - 21

EP - 25

BT - Green Cellular 2013

PB - Institute of Electrical and Electronic Engineers IEEE

ER -

Chen X, Zhang H, Lasanen M. Combined learning for energy efficiency in heterogeneous cellular networks. In Green Cellular 2013: 24th International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC Workshops 2013. Institute of Electrical and Electronic Engineers IEEE. 2013. p. 21-25 https://doi.org/10.1109/PIMRCW.2013.6707829