Tuning of Deep Neural Networks for Over-The-Air Linearization of Highly Nonlinear Wide-Band Active Phased Arrays

Feridoon Jalili; Yufeng Zhang; Felice Francesco Tafuri; Ole Kiel Jensen; Yunfeng Li; Qingyue Chen; Ming Shen; Gert Frølund Pedersen

doi:10.1109/ISNCC52172.2021.9615900

Tuning of Deep Neural Networks for Over-The-Air Linearization of Highly Nonlinear Wide-Band Active Phased Arrays

Feridoon Jalili^*, Yufeng Zhang, Felice Francesco Tafuri, Ole Kiel Jensen, Yunfeng Li, Qingyue Chen, Ming Shen, Gert Frølund Pedersen

^*Corresponding author for this work

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

1 Citation (Scopus)

Abstract

In this paper, we demonstrate how a deep neural network (DNN) can be used to compensate for nonlinearities and distortion effects introduced by the latest technology of 5G transmitters. A linearization approach based on neural networks can successfully cope with the challenges introduced not only by the high nonlinearity, wide bandwidth, and high frequency but also with challenges due to inter-PA crosstalk and load modulation. The device-under-test used in this experiment, is a state-of-the-art 5G 4x4 active phased array (APA) operating in very highly nonlinear regions at 28 GHz with a 100 MHz wide 3GPPbasestationsignalandwithOTAmeasuredsignalsusedfor training. Using the proposed DNN based linearization technique, an improvement of 11 % in error vector magnitude (EVM) and 10 dB adjacent channel leakage ratio (ACLR) are achieved which demonstrates the promising potential of this technique for emerging broadband communication systems such as 5G/6G and low earth orbit (LEO) satellite networks

Original language	English
Title of host publication	2021 International Symposium on Networks, Computers and Communications (ISNCC)
Number of pages	4
Publisher	IEEE
Publication date	10 Jun 2021
Pages	1-4
Article number	9615900
ISBN (Print)	978-1-6654-4829-1
ISBN (Electronic)	978-1-6654-0304-7
DOIs	https://doi.org/10.1109/ISNCC52172.2021.9615900
Publication status	Published - 10 Jun 2021
Event	2021 International Symposium on Networks, Computers and Communications (ISNCC) - Dubai, United Arab Emirates Duration: 31 Oct 2021 → 2 Nov 2021

Conference

Conference	2021 International Symposium on Networks, Computers and Communications (ISNCC)
Country/Territory	United Arab Emirates
City	Dubai
Period	31/10/2021 → 02/11/2021

Access to Document

10.1109/ISNCC52172.2021.9615900

AUB Link

Search for the material in Aalborg University Library's search engine

Cite this

Jalili, F., Zhang, Y., Tafuri, F. F., Jensen, O. K., Li, Y., Chen, Q., Shen, M., & Pedersen, G. F. (2021). Tuning of Deep Neural Networks for Over-The-Air Linearization of Highly Nonlinear Wide-Band Active Phased Arrays. In 2021 International Symposium on Networks, Computers and Communications (ISNCC) (pp. 1-4). Article 9615900 IEEE. https://doi.org/10.1109/ISNCC52172.2021.9615900

@inproceedings{d69db1357fc549cda637c49404aa1c29,

title = "Tuning of Deep Neural Networks for Over-The-Air Linearization of Highly Nonlinear Wide-Band Active Phased Arrays",

abstract = "In this paper, we demonstrate how a deep neural network (DNN) can be used to compensate for nonlinearities and distortion effects introduced by the latest technology of 5G transmitters. A linearization approach based on neural networks can successfully cope with the challenges introduced not only by the high nonlinearity, wide bandwidth, and high frequency but also with challenges due to inter-PA crosstalk and load modulation. The device-under-test used in this experiment, is a state-of-the-art 5G 4x4 active phased array (APA) operating in very highly nonlinear regions at 28 GHz with a 100 MHz wide 3GPPbasestationsignalandwithOTAmeasuredsignalsusedfor training. Using the proposed DNN based linearization technique, an improvement of 11 % in error vector magnitude (EVM) and 10 dB adjacent channel leakage ratio (ACLR) are achieved which demonstrates the promising potential of this technique for emerging broadband communication systems such as 5G/6G and low earth orbit (LEO) satellite networks",

author = "Feridoon Jalili and Yufeng Zhang and Tafuri, {Felice Francesco} and Jensen, {Ole Kiel} and Yunfeng Li and Qingyue Chen and Ming Shen and Pedersen, {Gert Fr{\o}lund}",

year = "2021",

month = jun,

day = "10",

doi = "10.1109/ISNCC52172.2021.9615900",

language = "English",

isbn = "978-1-6654-4829-1",

pages = "1--4",

booktitle = "2021 International Symposium on Networks, Computers and Communications (ISNCC)",

publisher = "IEEE",

address = "United States",

note = "2021 International Symposium on Networks, Computers and Communications (ISNCC) ; Conference date: 31-10-2021 Through 02-11-2021",

}

Jalili, F, Zhang, Y, Tafuri, FF, Jensen, OK, Li, Y, Chen, Q, Shen, M & Pedersen, GF 2021, Tuning of Deep Neural Networks for Over-The-Air Linearization of Highly Nonlinear Wide-Band Active Phased Arrays. in 2021 International Symposium on Networks, Computers and Communications (ISNCC)., 9615900, IEEE, pp. 1-4, 2021 International Symposium on Networks, Computers and Communications (ISNCC), Dubai, United Arab Emirates, 31/10/2021. https://doi.org/10.1109/ISNCC52172.2021.9615900

Tuning of Deep Neural Networks for Over-The-Air Linearization of Highly Nonlinear Wide-Band Active Phased Arrays. / Jalili, Feridoon; Zhang, Yufeng; Tafuri, Felice Francesco et al.
2021 International Symposium on Networks, Computers and Communications (ISNCC). IEEE, 2021. p. 1-4 9615900.

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

TY - GEN

T1 - Tuning of Deep Neural Networks for Over-The-Air Linearization of Highly Nonlinear Wide-Band Active Phased Arrays

AU - Jalili, Feridoon

AU - Zhang, Yufeng

AU - Tafuri, Felice Francesco

AU - Jensen, Ole Kiel

AU - Li, Yunfeng

AU - Chen, Qingyue

AU - Shen, Ming

AU - Pedersen, Gert Frølund

PY - 2021/6/10

Y1 - 2021/6/10

N2 - In this paper, we demonstrate how a deep neural network (DNN) can be used to compensate for nonlinearities and distortion effects introduced by the latest technology of 5G transmitters. A linearization approach based on neural networks can successfully cope with the challenges introduced not only by the high nonlinearity, wide bandwidth, and high frequency but also with challenges due to inter-PA crosstalk and load modulation. The device-under-test used in this experiment, is a state-of-the-art 5G 4x4 active phased array (APA) operating in very highly nonlinear regions at 28 GHz with a 100 MHz wide 3GPPbasestationsignalandwithOTAmeasuredsignalsusedfor training. Using the proposed DNN based linearization technique, an improvement of 11 % in error vector magnitude (EVM) and 10 dB adjacent channel leakage ratio (ACLR) are achieved which demonstrates the promising potential of this technique for emerging broadband communication systems such as 5G/6G and low earth orbit (LEO) satellite networks

AB - In this paper, we demonstrate how a deep neural network (DNN) can be used to compensate for nonlinearities and distortion effects introduced by the latest technology of 5G transmitters. A linearization approach based on neural networks can successfully cope with the challenges introduced not only by the high nonlinearity, wide bandwidth, and high frequency but also with challenges due to inter-PA crosstalk and load modulation. The device-under-test used in this experiment, is a state-of-the-art 5G 4x4 active phased array (APA) operating in very highly nonlinear regions at 28 GHz with a 100 MHz wide 3GPPbasestationsignalandwithOTAmeasuredsignalsusedfor training. Using the proposed DNN based linearization technique, an improvement of 11 % in error vector magnitude (EVM) and 10 dB adjacent channel leakage ratio (ACLR) are achieved which demonstrates the promising potential of this technique for emerging broadband communication systems such as 5G/6G and low earth orbit (LEO) satellite networks

U2 - 10.1109/ISNCC52172.2021.9615900

DO - 10.1109/ISNCC52172.2021.9615900

M3 - Article in proceeding

SN - 978-1-6654-4829-1

SP - 1

EP - 4

BT - 2021 International Symposium on Networks, Computers and Communications (ISNCC)

PB - IEEE

T2 - 2021 International Symposium on Networks, Computers and Communications (ISNCC)

Y2 - 31 October 2021 through 2 November 2021

ER -

Tuning of Deep Neural Networks for Over-The-Air Linearization of Highly Nonlinear Wide-Band Active Phased Arrays

Abstract

Conference

Access to Document

AUB Link

Fingerprint

Cite this