Feature Attention Based Blind Denoising Network for mmWave Beamspace Channel Estimation

Erhan Karakoca; Hasan Nayir; Ali Gorcin; Khalid Qaraqe

doi:10.1109/GCWkshps56602.2022.10008723

Feature Attention Based Blind Denoising Network for mmWave Beamspace Channel Estimation

Erhan Karakoca^*, Hasan Nayir, Ali Gorcin, Khalid Qaraqe

^*Corresponding author for this work

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

2 Citations (Scopus)

Abstract

In millimeter-wave (mmWave) MIMO systems, when the number of radio frequency (RF) chains are limited, estimation of the beamspace channel can become compelling. Also, as the number of RF chains decreases, pilot overhead increases to make channel estimation reliable, eventually reducing the spectral efficiency. In this paper, we propose a channel estimation method which combines compressive sensing (CS) method of GM-LAMP that assumes beamspace channel elements follows the Gaussian mixture distribution a priori, with a novel denoising network based on sparse feature attention for the estimation. According to performance analysis and simulation results, the GM-LAMP combined with feature attention based denoising neural network outperforms state-of-the-art compressed sensing-based algorithms. Furthermore, the proposed method also outperforms previous LAMP-based neural networks with comparable processing time, albeit using less pilot transmission.

Original language	English
Title of host publication	2022 IEEE GLOBECOM Workshops, GC Wkshps 2022 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1555-1560
Number of pages	6
ISBN (Electronic)	9781665459754
DOIs	https://doi.org/10.1109/GCWkshps56602.2022.10008723
Publication status	Published - 2022
Externally published	Yes
Event	2022 IEEE GLOBECOM Workshops, GC Wkshps 2022 - Virtual, Online, Brazil Duration: 4 Dec 2022 → 8 Dec 2022

Publication series

Name	2022 IEEE GLOBECOM Workshops, GC Wkshps 2022 - Proceedings

Conference

Conference	2022 IEEE GLOBECOM Workshops, GC Wkshps 2022
Country/Territory	Brazil
City	Virtual, Online
Period	4/12/22 → 8/12/22

Bibliographical note

Publisher Copyright:
© 2022 IEEE.

Keywords

AMP
Millimeter wave
beamspace MIMO
blind denoising
channel estimation
deep learning
neural networks
residual learning

Access to Document

10.1109/GCWkshps56602.2022.10008723

Cite this

Karakoca, E., Nayir, H., Gorcin, A., & Qaraqe, K. (2022). Feature Attention Based Blind Denoising Network for mmWave Beamspace Channel Estimation. In 2022 IEEE GLOBECOM Workshops, GC Wkshps 2022 - Proceedings (pp. 1555-1560). (2022 IEEE GLOBECOM Workshops, GC Wkshps 2022 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/GCWkshps56602.2022.10008723

@inproceedings{ba98a9b9e822466eb5f8a490ff527b61,

title = "Feature Attention Based Blind Denoising Network for mmWave Beamspace Channel Estimation",

abstract = "In millimeter-wave (mmWave) MIMO systems, when the number of radio frequency (RF) chains are limited, estimation of the beamspace channel can become compelling. Also, as the number of RF chains decreases, pilot overhead increases to make channel estimation reliable, eventually reducing the spectral efficiency. In this paper, we propose a channel estimation method which combines compressive sensing (CS) method of GM-LAMP that assumes beamspace channel elements follows the Gaussian mixture distribution a priori, with a novel denoising network based on sparse feature attention for the estimation. According to performance analysis and simulation results, the GM-LAMP combined with feature attention based denoising neural network outperforms state-of-the-art compressed sensing-based algorithms. Furthermore, the proposed method also outperforms previous LAMP-based neural networks with comparable processing time, albeit using less pilot transmission.",

keywords = "AMP, Millimeter wave, beamspace MIMO, blind denoising, channel estimation, deep learning, neural networks, residual learning",

author = "Erhan Karakoca and Hasan Nayir and Ali Gorcin and Khalid Qaraqe",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; 2022 IEEE GLOBECOM Workshops, GC Wkshps 2022 ; Conference date: 04-12-2022 Through 08-12-2022",

year = "2022",

doi = "10.1109/GCWkshps56602.2022.10008723",

language = "English",

series = "2022 IEEE GLOBECOM Workshops, GC Wkshps 2022 - Proceedings",

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

pages = "1555--1560",

booktitle = "2022 IEEE GLOBECOM Workshops, GC Wkshps 2022 - Proceedings",

address = "United States",

}

Karakoca, E, Nayir, H, Gorcin, A & Qaraqe, K 2022, Feature Attention Based Blind Denoising Network for mmWave Beamspace Channel Estimation. in 2022 IEEE GLOBECOM Workshops, GC Wkshps 2022 - Proceedings. 2022 IEEE GLOBECOM Workshops, GC Wkshps 2022 - Proceedings, Institute of Electrical and Electronics Engineers Inc., pp. 1555-1560, 2022 IEEE GLOBECOM Workshops, GC Wkshps 2022, Virtual, Online, Brazil, 4/12/22. https://doi.org/10.1109/GCWkshps56602.2022.10008723

Feature Attention Based Blind Denoising Network for mmWave Beamspace Channel Estimation. / Karakoca, Erhan; Nayir, Hasan; Gorcin, Ali et al.
2022 IEEE GLOBECOM Workshops, GC Wkshps 2022 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2022. p. 1555-1560 (2022 IEEE GLOBECOM Workshops, GC Wkshps 2022 - Proceedings).

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

TY - GEN

T1 - Feature Attention Based Blind Denoising Network for mmWave Beamspace Channel Estimation

AU - Karakoca, Erhan

AU - Nayir, Hasan

AU - Gorcin, Ali

AU - Qaraqe, Khalid

PY - 2022

Y1 - 2022

N2 - In millimeter-wave (mmWave) MIMO systems, when the number of radio frequency (RF) chains are limited, estimation of the beamspace channel can become compelling. Also, as the number of RF chains decreases, pilot overhead increases to make channel estimation reliable, eventually reducing the spectral efficiency. In this paper, we propose a channel estimation method which combines compressive sensing (CS) method of GM-LAMP that assumes beamspace channel elements follows the Gaussian mixture distribution a priori, with a novel denoising network based on sparse feature attention for the estimation. According to performance analysis and simulation results, the GM-LAMP combined with feature attention based denoising neural network outperforms state-of-the-art compressed sensing-based algorithms. Furthermore, the proposed method also outperforms previous LAMP-based neural networks with comparable processing time, albeit using less pilot transmission.

AB - In millimeter-wave (mmWave) MIMO systems, when the number of radio frequency (RF) chains are limited, estimation of the beamspace channel can become compelling. Also, as the number of RF chains decreases, pilot overhead increases to make channel estimation reliable, eventually reducing the spectral efficiency. In this paper, we propose a channel estimation method which combines compressive sensing (CS) method of GM-LAMP that assumes beamspace channel elements follows the Gaussian mixture distribution a priori, with a novel denoising network based on sparse feature attention for the estimation. According to performance analysis and simulation results, the GM-LAMP combined with feature attention based denoising neural network outperforms state-of-the-art compressed sensing-based algorithms. Furthermore, the proposed method also outperforms previous LAMP-based neural networks with comparable processing time, albeit using less pilot transmission.

KW - AMP

KW - Millimeter wave

KW - beamspace MIMO

KW - blind denoising

KW - channel estimation

KW - deep learning

KW - neural networks

KW - residual learning

UR - http://www.scopus.com/inward/record.url?scp=85146836066&partnerID=8YFLogxK

U2 - 10.1109/GCWkshps56602.2022.10008723

DO - 10.1109/GCWkshps56602.2022.10008723

M3 - Conference contribution

AN - SCOPUS:85146836066

T3 - 2022 IEEE GLOBECOM Workshops, GC Wkshps 2022 - Proceedings

SP - 1555

EP - 1560

BT - 2022 IEEE GLOBECOM Workshops, GC Wkshps 2022 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2022 IEEE GLOBECOM Workshops, GC Wkshps 2022

Y2 - 4 December 2022 through 8 December 2022

ER -

Karakoca E, Nayir H, Gorcin A, Qaraqe K. Feature Attention Based Blind Denoising Network for mmWave Beamspace Channel Estimation. In 2022 IEEE GLOBECOM Workshops, GC Wkshps 2022 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2022. p. 1555-1560. (2022 IEEE GLOBECOM Workshops, GC Wkshps 2022 - Proceedings). doi: 10.1109/GCWkshps56602.2022.10008723

Feature Attention Based Blind Denoising Network for mmWave Beamspace Channel Estimation

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this