A Calibrated Learning Approach to Distributed Power Allocation in Small Cell Networks

Xinruo Zhang; Mohammad Reza Nakhai; Gan Zheng; Sangarapillai Lambotharan; Bjorn Ottersten

A Calibrated Learning Approach to Distributed Power Allocation in Small Cell Networks

Xinruo Zhang, Mohammad Reza Nakhai, Gan Zheng, Sangarapillai Lambotharan, Bjorn Ottersten

Research output: Contribution to journal › Article › peer-review

40 Downloads (Pure)

Abstract

This paper studies the problem of max-min fairness power allocation in distributed small cell networks operated under the same frequency bandwidth. We introduce a calibrated learning algorithm to account for the dynamism in wireless channels and achieve max-min user fairness in the long run. Provided that the small base stations (SBSs) serve their own users simultaneously and no prior knowledge of forthcoming user demand is available at the SBSs, the aim of the proposed algorithm is to allow SBSs to gradually improve their fore- cast of the possible transmit power levels of other SBSs and make best decisions based on the predicted results at indi- vidual time slots. Simulation results validate that in terms of achieving max-min signal-to-interference-plus-noise ratio, the proposed distributed design outperforms two benchmark schemes and achieves a similar performance as compared to the optimal centralized design.

Original language	English
Journal	2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2019)
Publication status	Accepted/In press - 1 Feb 2019

Access to Document

ICASSP_xinruo_4pagesAccepted author manuscript, 254 KB

Cite this

@article{6a5c43ffff8b46cf8754e4e5ceaab9a6,

title = "A Calibrated Learning Approach to Distributed Power Allocation in Small Cell Networks",

abstract = "This paper studies the problem of max-min fairness power allocation in distributed small cell networks operated under the same frequency bandwidth. We introduce a calibrated learning algorithm to account for the dynamism in wireless channels and achieve max-min user fairness in the long run. Provided that the small base stations (SBSs) serve their own users simultaneously and no prior knowledge of forthcoming user demand is available at the SBSs, the aim of the proposed algorithm is to allow SBSs to gradually improve their fore- cast of the possible transmit power levels of other SBSs and make best decisions based on the predicted results at indi- vidual time slots. Simulation results validate that in terms of achieving max-min signal-to-interference-plus-noise ratio, the proposed distributed design outperforms two benchmark schemes and achieves a similar performance as compared to the optimal centralized design.",

author = "Xinruo Zhang and Nakhai, {Mohammad Reza} and Gan Zheng and Sangarapillai Lambotharan and Bjorn Ottersten",

year = "2019",

month = feb,

day = "1",

language = "English",

journal = "2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2019)",

publisher = "IEEE",

}

TY - JOUR

T1 - A Calibrated Learning Approach to Distributed Power Allocation in Small Cell Networks

AU - Zhang, Xinruo

AU - Nakhai, Mohammad Reza

AU - Zheng, Gan

AU - Lambotharan, Sangarapillai

AU - Ottersten, Bjorn

PY - 2019/2/1

Y1 - 2019/2/1

N2 - This paper studies the problem of max-min fairness power allocation in distributed small cell networks operated under the same frequency bandwidth. We introduce a calibrated learning algorithm to account for the dynamism in wireless channels and achieve max-min user fairness in the long run. Provided that the small base stations (SBSs) serve their own users simultaneously and no prior knowledge of forthcoming user demand is available at the SBSs, the aim of the proposed algorithm is to allow SBSs to gradually improve their fore- cast of the possible transmit power levels of other SBSs and make best decisions based on the predicted results at indi- vidual time slots. Simulation results validate that in terms of achieving max-min signal-to-interference-plus-noise ratio, the proposed distributed design outperforms two benchmark schemes and achieves a similar performance as compared to the optimal centralized design.

AB - This paper studies the problem of max-min fairness power allocation in distributed small cell networks operated under the same frequency bandwidth. We introduce a calibrated learning algorithm to account for the dynamism in wireless channels and achieve max-min user fairness in the long run. Provided that the small base stations (SBSs) serve their own users simultaneously and no prior knowledge of forthcoming user demand is available at the SBSs, the aim of the proposed algorithm is to allow SBSs to gradually improve their fore- cast of the possible transmit power levels of other SBSs and make best decisions based on the predicted results at indi- vidual time slots. Simulation results validate that in terms of achieving max-min signal-to-interference-plus-noise ratio, the proposed distributed design outperforms two benchmark schemes and achieves a similar performance as compared to the optimal centralized design.

M3 - Article

JO - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2019)

JF - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2019)

ER -

A Calibrated Learning Approach to Distributed Power Allocation in Small Cell Networks

Abstract

Access to Document

Fingerprint

Cite this