Energy minimization for federated learning with IRS-assisted over-the-air computation

Yuntao Hu; Ming Chen; Mingzhe Chen; Zhaohui Yang; Mohammad Shikh-Bahaei; H. Vincent Poor; Shuguang Cui

doi:10.1109/ICASSP39728.2021.9414785

Energy minimization for federated learning with IRS-assisted over-the-air computation

Yuntao Hu, Ming Chen, Mingzhe Chen, Zhaohui Yang, Mohammad Shikh-Bahaei, H. Vincent Poor, Shuguang Cui

Research output: Contribution to journal › Conference paper › peer-review

18 Citations (Scopus)

Abstract

This paper investigates the deployment of federated learning (FL) over an over-the-air computation (AirComp) and intelligent reflecting surface (IRS) based wireless network. In the considered system, devices transmit locally trained machine learning (ML) models to the base station (BS) which aggregates the received ML models and generates a shared global ML model. The devices can directly transmit ML models to the BS or using IRS. Meanwhile, AirComp is used to aggregate ML models that are transmitted from the devices to the BS. To minimize the energy consumption of devices, an energy minimization problem is formulated, which jointly optimizes the device selection, phase shift matrix, decoding vector, and power control. To seek the solution, the original optimization problem is divided into four sub-problems. Then the fractional program, greedy algorithm, matrix derivation, and weighted minimum mean square error methods are used to compute the phase shift matrix, device selection vector, decoding vector, and transmit power, respectively. Simulation results show that the proposed algorithm can reduce 11.2% energy consumption of devices compared to an FL algorithm that is implemented at a network without any IRSs.

Original language	English
Pages (from-to)	3105-3109
Number of pages	5
Journal	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume	2021-June
DOIs	https://doi.org/10.1109/ICASSP39728.2021.9414785
Publication status	Published - 2021
Event	2021 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2021 - Virtual, Toronto, Canada Duration: 6 Jun 2021 → 11 Jun 2021

Access to Document

10.1109/ICASSP39728.2021.9414785

Cite this

@article{6939003c1dc64f179714129c4658afde,

title = "Energy minimization for federated learning with IRS-assisted over-the-air computation",

abstract = "This paper investigates the deployment of federated learning (FL) over an over-the-air computation (AirComp) and intelligent reflecting surface (IRS) based wireless network. In the considered system, devices transmit locally trained machine learning (ML) models to the base station (BS) which aggregates the received ML models and generates a shared global ML model. The devices can directly transmit ML models to the BS or using IRS. Meanwhile, AirComp is used to aggregate ML models that are transmitted from the devices to the BS. To minimize the energy consumption of devices, an energy minimization problem is formulated, which jointly optimizes the device selection, phase shift matrix, decoding vector, and power control. To seek the solution, the original optimization problem is divided into four sub-problems. Then the fractional program, greedy algorithm, matrix derivation, and weighted minimum mean square error methods are used to compute the phase shift matrix, device selection vector, decoding vector, and transmit power, respectively. Simulation results show that the proposed algorithm can reduce 11.2% energy consumption of devices compared to an FL algorithm that is implemented at a network without any IRSs.",

author = "Yuntao Hu and Ming Chen and Mingzhe Chen and Zhaohui Yang and Mohammad Shikh-Bahaei and Poor, {H. Vincent} and Shuguang Cui",

note = "Funding Information: We gratefully acknowledge the support of this work from the National Natural Science Foundation of China (NSFC) under grant 61871128 and U.S. National Science Foundation under grant CCF-1908308. Y. Hu and M. Chen is with the National Mobile Communications Research Laboratory, South-east University, Nanjing, China (email: {huyuntao, chenming}@seu.edu.cn). Mingzhe Chen and H. V. Poor are with the EE Dept., Princeton University, Princeton, NJ (e-mail: {mingzhec, poor}@princeton.edu). Z. Yang and Mohammad Shikh-Bahaei are with Centre for TelecommunicationsResearch, Department of Engineering, King{\textquoteright}s College London (email: {yang.zhaohui, m.sbahaei}@kcl.ac.uk). S. Cui is with Shenzhen Research Institute of Big Data and School of Science and Engineering, the Chinese University of Hong Kong, China (e-mail: [email protected]) Publisher Copyright: {\textcopyright} 2021 IEEE Copyright: Copyright 2021 Elsevier B.V., All rights reserved.; 2021 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2021 ; Conference date: 06-06-2021 Through 11-06-2021",

year = "2021",

doi = "10.1109/ICASSP39728.2021.9414785",

language = "English",

volume = "2021-June",

pages = "3105--3109",

journal = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

issn = "1520-6149",

}

TY - JOUR

T1 - Energy minimization for federated learning with IRS-assisted over-the-air computation

AU - Hu, Yuntao

AU - Chen, Ming

AU - Chen, Mingzhe

AU - Yang, Zhaohui

AU - Shikh-Bahaei, Mohammad

AU - Poor, H. Vincent

AU - Cui, Shuguang

N1 - Funding Information: We gratefully acknowledge the support of this work from the National Natural Science Foundation of China (NSFC) under grant 61871128 and U.S. National Science Foundation under grant CCF-1908308. Y. Hu and M. Chen is with the National Mobile Communications Research Laboratory, South-east University, Nanjing, China (email: {huyuntao, chenming}@seu.edu.cn). Mingzhe Chen and H. V. Poor are with the EE Dept., Princeton University, Princeton, NJ (e-mail: {mingzhec, poor}@princeton.edu). Z. Yang and Mohammad Shikh-Bahaei are with Centre for TelecommunicationsResearch, Department of Engineering, King’s College London (email: {yang.zhaohui, m.sbahaei}@kcl.ac.uk). S. Cui is with Shenzhen Research Institute of Big Data and School of Science and Engineering, the Chinese University of Hong Kong, China (e-mail: [email protected]) Publisher Copyright: © 2021 IEEE Copyright: Copyright 2021 Elsevier B.V., All rights reserved.

PY - 2021

Y1 - 2021

N2 - This paper investigates the deployment of federated learning (FL) over an over-the-air computation (AirComp) and intelligent reflecting surface (IRS) based wireless network. In the considered system, devices transmit locally trained machine learning (ML) models to the base station (BS) which aggregates the received ML models and generates a shared global ML model. The devices can directly transmit ML models to the BS or using IRS. Meanwhile, AirComp is used to aggregate ML models that are transmitted from the devices to the BS. To minimize the energy consumption of devices, an energy minimization problem is formulated, which jointly optimizes the device selection, phase shift matrix, decoding vector, and power control. To seek the solution, the original optimization problem is divided into four sub-problems. Then the fractional program, greedy algorithm, matrix derivation, and weighted minimum mean square error methods are used to compute the phase shift matrix, device selection vector, decoding vector, and transmit power, respectively. Simulation results show that the proposed algorithm can reduce 11.2% energy consumption of devices compared to an FL algorithm that is implemented at a network without any IRSs.

AB - This paper investigates the deployment of federated learning (FL) over an over-the-air computation (AirComp) and intelligent reflecting surface (IRS) based wireless network. In the considered system, devices transmit locally trained machine learning (ML) models to the base station (BS) which aggregates the received ML models and generates a shared global ML model. The devices can directly transmit ML models to the BS or using IRS. Meanwhile, AirComp is used to aggregate ML models that are transmitted from the devices to the BS. To minimize the energy consumption of devices, an energy minimization problem is formulated, which jointly optimizes the device selection, phase shift matrix, decoding vector, and power control. To seek the solution, the original optimization problem is divided into four sub-problems. Then the fractional program, greedy algorithm, matrix derivation, and weighted minimum mean square error methods are used to compute the phase shift matrix, device selection vector, decoding vector, and transmit power, respectively. Simulation results show that the proposed algorithm can reduce 11.2% energy consumption of devices compared to an FL algorithm that is implemented at a network without any IRSs.

UR - http://www.scopus.com/inward/record.url?scp=85115111616&partnerID=8YFLogxK

U2 - 10.1109/ICASSP39728.2021.9414785

DO - 10.1109/ICASSP39728.2021.9414785

M3 - Conference paper

AN - SCOPUS:85115111616

SN - 1520-6149

VL - 2021-June

SP - 3105

EP - 3109

JO - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

JF - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

T2 - 2021 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2021

Y2 - 6 June 2021 through 11 June 2021

ER -

Energy minimization for federated learning with IRS-assisted over-the-air computation

Abstract

Access to Document

Other files and links

Fingerprint

Cite this