Antenna Array Beamforming Based on Deep Learning Neural Network Architectures

Haya Al Kassir; Zaharias D. Zaharis; Pavlos I. Lazaridis; Nikolaos V. Kantartzis; Traianos V. Yioultsis; Ioannis P. Chochliouros; Albena Mihovska; Thomas D. Xenos

doi:10.23919/AT-AP-RASC54737.2022.9814201

Antenna Array Beamforming Based on Deep Learning Neural Network Architectures

Haya Al Kassir, Zaharias D. Zaharis, Pavlos I. Lazaridis, Nikolaos V. Kantartzis, Traianos V. Yioultsis, Ioannis P. Chochliouros, Albena Mihovska, Thomas D. Xenos

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

14 Citations (Scopus)

Abstract

The implementation of antenna array beamforming using several neural network (NN) architectures is compared in this paper. Gated recurrent unit, feed-forward NN, convolutional NN, and long short-term memory architectures have been used for the beamforming process. This comparative study is carried out using various metrics, such as the root mean square error, and the computational time for each NN. In addition, the mean absolute divergences of the antenna array main lobe and nulls directions from their respective desired directions have also been used to assess the performance of each beamformer. The neural networks are trained using the simulation results of a 16-element microstrip patch antenna array. It is demonstrated that deep learning-based beamformers are capable of computing optimum antenna array weights in real time and in environments that change with time.

Original language	English
Title of host publication	2022 3rd URSI Atlantic and Asia Pacific Radio Science Meeting, AT-AP-RASC 2022
Publisher	Institute of Electrical and Electronics Engineers Inc.
Number of pages	4
ISBN (Electronic)	9789463968058
ISBN (Print)	9781665499866
DOIs	https://doi.org/10.23919/AT-AP-RASC54737.2022.9814201
Publication status	Published - 6 Jul 2022
Event	3rd URSI Atlantic and Asia Pacific Radio Science Meeting - Gran Canaria, Spain Duration: 29 May 2022 → 3 Jun 2022 Conference number: 3

Conference

Conference	3rd URSI Atlantic and Asia Pacific Radio Science Meeting
Abbreviated title	AT-AP-RASC 2022
Country/Territory	Spain
City	Gran Canaria
Period	29/05/22 → 3/06/22

UN SDGs

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

SDG 9 Industry, Innovation, and Infrastructure

Access to Document

10.23919/AT-AP-RASC54737.2022.9814201Licence: Unspecified

Accepted manuscriptAccepted author manuscript, 260 KBLicence: Unspecified

Cite this

Al Kassir, H., Zaharis, Z. D., Lazaridis, P. I., Kantartzis, N. V., Yioultsis, T. V., Chochliouros, I. P., Mihovska, A., & Xenos, T. D. (2022). Antenna Array Beamforming Based on Deep Learning Neural Network Architectures. In 2022 3rd URSI Atlantic and Asia Pacific Radio Science Meeting, AT-AP-RASC 2022 Article 9814201 Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.23919/AT-AP-RASC54737.2022.9814201

@inproceedings{55fc5c4435a644d2831034181ec09bcf,

title = "Antenna Array Beamforming Based on Deep Learning Neural Network Architectures",

abstract = "The implementation of antenna array beamforming using several neural network (NN) architectures is compared in this paper. Gated recurrent unit, feed-forward NN, convolutional NN, and long short-term memory architectures have been used for the beamforming process. This comparative study is carried out using various metrics, such as the root mean square error, and the computational time for each NN. In addition, the mean absolute divergences of the antenna array main lobe and nulls directions from their respective desired directions have also been used to assess the performance of each beamformer. The neural networks are trained using the simulation results of a 16-element microstrip patch antenna array. It is demonstrated that deep learning-based beamformers are capable of computing optimum antenna array weights in real time and in environments that change with time. ",

keywords = "Deep learning, Microstrip antenna arrays, Array signal processing, Simulation, Patch antennas, Artificial neural networks, Real-time systems",

author = "\{Al Kassir\}, Haya and Zaharis, \{Zaharias D.\} and Lazaridis, \{Pavlos I.\} and Kantartzis, \{Nikolaos V.\} and Yioultsis, \{Traianos V.\} and Chochliouros, \{Ioannis P.\} and Albena Mihovska and Xenos, \{Thomas D.\}",

note = "Funding Information: This research was supported by the European Union, partially through the Horizon 2020 Marie Sk{\l}odowska-Curie Innovative Training Networks Programme “Mobility and Training for beyond 5G Ecosystems (MOTOR5G)” under grant agreement no. 861219, and partially through the Horizon 2020 Marie Sk{\l}odowska-Curie Research and Innovation Staff Exchange Programme “Research Collaboration and Mobility for Beyond 5G Future Wireless Networks (RECOMBINE)” under grant agreement no. 872857. Publisher Copyright: {\textcopyright} 2022 URSI.; 3rd URSI Atlantic and Asia Pacific Radio Science Meeting, AT-AP-RASC 2022 ; Conference date: 29-05-2022 Through 03-06-2022",

year = "2022",

month = jul,

day = "6",

doi = "10.23919/AT-AP-RASC54737.2022.9814201",

language = "English",

isbn = "9781665499866",

booktitle = "2022 3rd URSI Atlantic and Asia Pacific Radio Science Meeting, AT-AP-RASC 2022",

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

address = "United States",

}

Al Kassir, H, Zaharis, ZD, Lazaridis, PI, Kantartzis, NV, Yioultsis, TV, Chochliouros, IP, Mihovska, A & Xenos, TD 2022, Antenna Array Beamforming Based on Deep Learning Neural Network Architectures. in 2022 3rd URSI Atlantic and Asia Pacific Radio Science Meeting, AT-AP-RASC 2022., 9814201, Institute of Electrical and Electronics Engineers Inc., 3rd URSI Atlantic and Asia Pacific Radio Science Meeting, Gran Canaria, Spain, 29/05/22. https://doi.org/10.23919/AT-AP-RASC54737.2022.9814201

Antenna Array Beamforming Based on Deep Learning Neural Network Architectures. / Al Kassir, Haya; Zaharis, Zaharias D.; Lazaridis, Pavlos I. et al.
2022 3rd URSI Atlantic and Asia Pacific Radio Science Meeting, AT-AP-RASC 2022. Institute of Electrical and Electronics Engineers Inc., 2022. 9814201.

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

TY - GEN

T1 - Antenna Array Beamforming Based on Deep Learning Neural Network Architectures

AU - Al Kassir, Haya

AU - Zaharis, Zaharias D.

AU - Lazaridis, Pavlos I.

AU - Kantartzis, Nikolaos V.

AU - Yioultsis, Traianos V.

AU - Chochliouros, Ioannis P.

AU - Mihovska, Albena

AU - Xenos, Thomas D.

N1 - Conference code: 3

PY - 2022/7/6

Y1 - 2022/7/6

N2 - The implementation of antenna array beamforming using several neural network (NN) architectures is compared in this paper. Gated recurrent unit, feed-forward NN, convolutional NN, and long short-term memory architectures have been used for the beamforming process. This comparative study is carried out using various metrics, such as the root mean square error, and the computational time for each NN. In addition, the mean absolute divergences of the antenna array main lobe and nulls directions from their respective desired directions have also been used to assess the performance of each beamformer. The neural networks are trained using the simulation results of a 16-element microstrip patch antenna array. It is demonstrated that deep learning-based beamformers are capable of computing optimum antenna array weights in real time and in environments that change with time.

AB - The implementation of antenna array beamforming using several neural network (NN) architectures is compared in this paper. Gated recurrent unit, feed-forward NN, convolutional NN, and long short-term memory architectures have been used for the beamforming process. This comparative study is carried out using various metrics, such as the root mean square error, and the computational time for each NN. In addition, the mean absolute divergences of the antenna array main lobe and nulls directions from their respective desired directions have also been used to assess the performance of each beamformer. The neural networks are trained using the simulation results of a 16-element microstrip patch antenna array. It is demonstrated that deep learning-based beamformers are capable of computing optimum antenna array weights in real time and in environments that change with time.

KW - Deep learning

KW - Microstrip antenna arrays

KW - Array signal processing

KW - Simulation

KW - Patch antennas

KW - Artificial neural networks

KW - Real-time systems

UR - https://www.scopus.com/pages/publications/85134881611

U2 - 10.23919/AT-AP-RASC54737.2022.9814201

DO - 10.23919/AT-AP-RASC54737.2022.9814201

M3 - Conference contribution

AN - SCOPUS:85134881611

SN - 9781665499866

BT - 2022 3rd URSI Atlantic and Asia Pacific Radio Science Meeting, AT-AP-RASC 2022

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 3rd URSI Atlantic and Asia Pacific Radio Science Meeting

Y2 - 29 May 2022 through 3 June 2022

ER -

Antenna Array Beamforming Based on Deep Learning Neural Network Architectures

Abstract

Conference

UN SDGs

Access to Document

Other files and links

Fingerprint

MOTOR5G: H2020 ITN MOTOR5G : MObility and Training fOR beyond 5G Ecosystems

H2020 RISE RECOMBINE: Research Collaboration and Mobility for Beyond 5G Future Wireless Networks

Cite this

Antenna Array Beamforming Based on Deep Learning Neural Network Architectures

Abstract

Conference

UN SDGs

Access to Document

Other files and links

Fingerprint

Projects

MOTOR5G: H2020 ITN MOTOR5G : MObility and Training fOR beyond 5G Ecosystems

H2020 RISE RECOMBINE: Research Collaboration and Mobility for Beyond 5G Future Wireless Networks

Cite this