Power Control in massive MIMO Networks Using Transfer Learning with Deep Neural Networks

Neda Ahmadi; Iosif Mporas; Anastasios Papazafeiropoulos; Pandelis Kourtessis; John Senior

doi:10.1109/CAMAD55695.2022.9966903

Power Control in massive MIMO Networks Using Transfer Learning with Deep Neural Networks

Neda Ahmadi, Iosif Mporas, Anastasios Papazafeiropoulos, Pandelis Kourtessis, John Senior

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

Abstract

Power control (PC) plays a crucial role in massive multiple-input-multiple-output (mMIMO) networks. There are several heuristic algorithms, like the weighted mean square error (WMMSE) algorithm, used to optimise the PC. In order these algorithms to perform the power allocation they require high computational power. In this paper, we address this problem through the application of machine learning (ML)-based algorithms as they can produce close to optimal solutions with a very low computational complexity. We propose the use of transfer learning with deep neural networks (TLDNN) under the objective of maximising the sum spectral efficiency (SE). The evaluation results demonstrate that the TLDNN approach outperforms the deep neural network (DNN) based PC and is twice faster than the WMMSE based PC.

Original language	English
Title of host publication	2022 IEEE 27th International Workshop on Computer Aided Modeling and Design of Communication Links and Networks, CAMAD 2022
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Pages	89-93
Number of pages	5
ISBN (Electronic)	9781665461290
DOIs	https://doi.org/10.1109/CAMAD55695.2022.9966903
Publication status	Published - 2022
Externally published	Yes
Event	27th IEEE International Workshop on Computer Aided Modeling and Design of Communication Links and Networks, CAMAD 2022 - Paris, France Duration: 2 Nov 2022 → 3 Nov 2022

Publication series

Name	IEEE International Workshop on Computer Aided Modeling and Design of Communication Links and Networks, CAMAD
Volume	2022-November
ISSN (Electronic)	2378-4873

Conference

Conference	27th IEEE International Workshop on Computer Aided Modeling and Design of Communication Links and Networks, CAMAD 2022
Country/Territory	France
City	Paris
Period	2/11/22 → 3/11/22

Keywords

deep neural network
machine learning
massive MIMO
power control
transfer learning

Access to Document

10.1109/CAMAD55695.2022.9966903

Cite this

Ahmadi, N., Mporas, I., Papazafeiropoulos, A., Kourtessis, P., & Senior, J. (2022). Power Control in massive MIMO Networks Using Transfer Learning with Deep Neural Networks. In 2022 IEEE 27th International Workshop on Computer Aided Modeling and Design of Communication Links and Networks, CAMAD 2022 (pp. 89-93). (IEEE International Workshop on Computer Aided Modeling and Design of Communication Links and Networks, CAMAD; Vol. 2022-November). Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/CAMAD55695.2022.9966903

Ahmadi, Neda ; Mporas, Iosif ; Papazafeiropoulos, Anastasios et al. / Power Control in massive MIMO Networks Using Transfer Learning with Deep Neural Networks. 2022 IEEE 27th International Workshop on Computer Aided Modeling and Design of Communication Links and Networks, CAMAD 2022. Institute of Electrical and Electronics Engineers (IEEE), 2022. pp. 89-93 (IEEE International Workshop on Computer Aided Modeling and Design of Communication Links and Networks, CAMAD).

@inproceedings{83ffb44e30e349f0ae5710d4669316fe,

title = "Power Control in massive MIMO Networks Using Transfer Learning with Deep Neural Networks",

abstract = "Power control (PC) plays a crucial role in massive multiple-input-multiple-output (mMIMO) networks. There are several heuristic algorithms, like the weighted mean square error (WMMSE) algorithm, used to optimise the PC. In order these algorithms to perform the power allocation they require high computational power. In this paper, we address this problem through the application of machine learning (ML)-based algorithms as they can produce close to optimal solutions with a very low computational complexity. We propose the use of transfer learning with deep neural networks (TLDNN) under the objective of maximising the sum spectral efficiency (SE). The evaluation results demonstrate that the TLDNN approach outperforms the deep neural network (DNN) based PC and is twice faster than the WMMSE based PC.",

keywords = "deep neural network, machine learning, massive MIMO, power control, transfer learning",

author = "Neda Ahmadi and Iosif Mporas and Anastasios Papazafeiropoulos and Pandelis Kourtessis and John Senior",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; 27th IEEE International Workshop on Computer Aided Modeling and Design of Communication Links and Networks, CAMAD 2022 ; Conference date: 02-11-2022 Through 03-11-2022",

year = "2022",

doi = "10.1109/CAMAD55695.2022.9966903",

language = "English",

series = "IEEE International Workshop on Computer Aided Modeling and Design of Communication Links and Networks, CAMAD",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

pages = "89--93",

booktitle = "2022 IEEE 27th International Workshop on Computer Aided Modeling and Design of Communication Links and Networks, CAMAD 2022",

}

Ahmadi, N, Mporas, I , Papazafeiropoulos, A , Kourtessis, P & Senior, J 2022, Power Control in massive MIMO Networks Using Transfer Learning with Deep Neural Networks. in 2022 IEEE 27th International Workshop on Computer Aided Modeling and Design of Communication Links and Networks, CAMAD 2022. IEEE International Workshop on Computer Aided Modeling and Design of Communication Links and Networks, CAMAD, vol. 2022-November, Institute of Electrical and Electronics Engineers (IEEE), pp. 89-93, 27th IEEE International Workshop on Computer Aided Modeling and Design of Communication Links and Networks, CAMAD 2022, Paris, France, 2/11/22. https://doi.org/10.1109/CAMAD55695.2022.9966903

Power Control in massive MIMO Networks Using Transfer Learning with Deep Neural Networks. / Ahmadi, Neda; Mporas, Iosif ; Papazafeiropoulos, Anastasios et al.

2022 IEEE 27th International Workshop on Computer Aided Modeling and Design of Communication Links and Networks, CAMAD 2022. Institute of Electrical and Electronics Engineers (IEEE), 2022. p. 89-93 (IEEE International Workshop on Computer Aided Modeling and Design of Communication Links and Networks, CAMAD; Vol. 2022-November).

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

TY - GEN

T1 - Power Control in massive MIMO Networks Using Transfer Learning with Deep Neural Networks

AU - Ahmadi, Neda

AU - Mporas, Iosif

AU - Papazafeiropoulos, Anastasios

AU - Kourtessis, Pandelis

AU - Senior, John

PY - 2022

Y1 - 2022

N2 - Power control (PC) plays a crucial role in massive multiple-input-multiple-output (mMIMO) networks. There are several heuristic algorithms, like the weighted mean square error (WMMSE) algorithm, used to optimise the PC. In order these algorithms to perform the power allocation they require high computational power. In this paper, we address this problem through the application of machine learning (ML)-based algorithms as they can produce close to optimal solutions with a very low computational complexity. We propose the use of transfer learning with deep neural networks (TLDNN) under the objective of maximising the sum spectral efficiency (SE). The evaluation results demonstrate that the TLDNN approach outperforms the deep neural network (DNN) based PC and is twice faster than the WMMSE based PC.

AB - Power control (PC) plays a crucial role in massive multiple-input-multiple-output (mMIMO) networks. There are several heuristic algorithms, like the weighted mean square error (WMMSE) algorithm, used to optimise the PC. In order these algorithms to perform the power allocation they require high computational power. In this paper, we address this problem through the application of machine learning (ML)-based algorithms as they can produce close to optimal solutions with a very low computational complexity. We propose the use of transfer learning with deep neural networks (TLDNN) under the objective of maximising the sum spectral efficiency (SE). The evaluation results demonstrate that the TLDNN approach outperforms the deep neural network (DNN) based PC and is twice faster than the WMMSE based PC.

KW - deep neural network

KW - machine learning

KW - massive MIMO

KW - power control

KW - transfer learning

UR - http://www.scopus.com/inward/record.url?scp=85144035004&partnerID=8YFLogxK

U2 - 10.1109/CAMAD55695.2022.9966903

DO - 10.1109/CAMAD55695.2022.9966903

M3 - Conference contribution

AN - SCOPUS:85144035004

T3 - IEEE International Workshop on Computer Aided Modeling and Design of Communication Links and Networks, CAMAD

SP - 89

EP - 93

BT - 2022 IEEE 27th International Workshop on Computer Aided Modeling and Design of Communication Links and Networks, CAMAD 2022

PB - Institute of Electrical and Electronics Engineers (IEEE)

T2 - 27th IEEE International Workshop on Computer Aided Modeling and Design of Communication Links and Networks, CAMAD 2022

Y2 - 2 November 2022 through 3 November 2022

ER -

Ahmadi N, Mporas I , Papazafeiropoulos A , Kourtessis P , Senior J. Power Control in massive MIMO Networks Using Transfer Learning with Deep Neural Networks. In 2022 IEEE 27th International Workshop on Computer Aided Modeling and Design of Communication Links and Networks, CAMAD 2022. Institute of Electrical and Electronics Engineers (IEEE). 2022. p. 89-93. (IEEE International Workshop on Computer Aided Modeling and Design of Communication Links and Networks, CAMAD). doi: 10.1109/CAMAD55695.2022.9966903

Power Control in massive MIMO Networks Using Transfer Learning with Deep Neural Networks

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this