Model-free learning of optimal beamformers for passive IRS-assisted sumrate maximization

Hassaan Hashmi; Spyridon Pougkakiotis; Dionysis S. Kalogerias

doi:10.1109/ICASSP49357.2023.10095269

Model-free learning of optimal beamformers for passive IRS-assisted sumrate maximization

Hassaan Hashmi, Spyridon Pougkakiotis, Dionysis S. Kalogerias

Mathematics

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

2 Citations (Scopus)

Abstract

Although Intelligent Reflective Surfaces (IRSs) are a cost-effective technology promising high spectral efficiency in future wireless networks, obtaining optimal IRS beamformers is a challenging problem with several practical limitations. Assuming fully-passive, sensing- free IRS operation, we introduce a new data-driven Zeroth-order Stochastic Gradient Ascent (ZoSGA) algorithm for sumrate optimization in an IRS-aided downlink setting. ZoSGA does not require access to channel model or network structure information, and enables learning of optimal long-term IRS beamformers jointly with standard short-term precoding, based only on conventional effective channel state information. Supported by state-of-the-art (SOTA) convergence analysis, detailed simulations confirm that ZoSGA exhibits SOTA empirical behavior as well, consistently outperforming standard fully model-based baselines, in a variety of scenarios.

Original language	English
Title of host publication	ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)
Publisher	IEEE
Pages	1-5
Number of pages	5
ISBN (Electronic)	9781728163277
ISBN (Print)	9781728163284
DOIs	https://doi.org/10.1109/ICASSP49357.2023.10095269
Publication status	Published - 10 Jun 2023
Event	ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) - Rhodes Island, Greece, Greece Duration: 4 Jun 2023 → 10 Jun 2023 https://2023.ieeeicassp.org/

Publication series

Name	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume	2023-June
ISSN (Print)	1520-6149

Conference

Conference	ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Abbreviated title	ICASSP 2023
Country/Territory	Greece
Period	4/06/23 → 10/06/23
Internet address	https://2023.ieeeicassp.org/

Keywords

Intelligent Reflecting Surfaces
Model-Free Learning
Sumrate Maximization
Zeroth-order Optimization

ASJC Scopus subject areas

Software
Signal Processing
Electrical and Electronic Engineering

Access to Document

10.1109/ICASSP49357.2023.10095269

Cite this

Hashmi, H., Pougkakiotis, S., & Kalogerias, D. S. (2023). Model-free learning of optimal beamformers for passive IRS-assisted sumrate maximization. In ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP) (pp. 1-5). Article 10095269 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2023-June). IEEE. https://doi.org/10.1109/ICASSP49357.2023.10095269

Hashmi, Hassaan ; Pougkakiotis, Spyridon ; Kalogerias, Dionysis S. / Model-free learning of optimal beamformers for passive IRS-assisted sumrate maximization. ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP). IEEE, 2023. pp. 1-5 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

@inproceedings{f10f30777cec4ee58e4c2644db81610d,

title = "Model-free learning of optimal beamformers for passive IRS-assisted sumrate maximization",

abstract = "Although Intelligent Reflective Surfaces (IRSs) are a cost-effective technology promising high spectral efficiency in future wireless networks, obtaining optimal IRS beamformers is a challenging problem with several practical limitations. Assuming fully-passive, sensing- free IRS operation, we introduce a new data-driven Zeroth-order Stochastic Gradient Ascent (ZoSGA) algorithm for sumrate optimization in an IRS-aided downlink setting. ZoSGA does not require access to channel model or network structure information, and enables learning of optimal long-term IRS beamformers jointly with standard short-term precoding, based only on conventional effective channel state information. Supported by state-of-the-art (SOTA) convergence analysis, detailed simulations confirm that ZoSGA exhibits SOTA empirical behavior as well, consistently outperforming standard fully model-based baselines, in a variety of scenarios.",

keywords = "Intelligent Reflecting Surfaces, Model-Free Learning, Sumrate Maximization, Zeroth-order Optimization",

author = "Hassaan Hashmi and Spyridon Pougkakiotis and Kalogerias, {Dionysis S.}",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), ICASSP 2023 ; Conference date: 04-06-2023 Through 10-06-2023",

year = "2023",

month = jun,

day = "10",

doi = "10.1109/ICASSP49357.2023.10095269",

language = "English",

isbn = "9781728163284",

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

publisher = "IEEE",

pages = "1--5",

booktitle = "ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)",

url = "https://2023.ieeeicassp.org/",

}

Hashmi, H, Pougkakiotis, S & Kalogerias, DS 2023, Model-free learning of optimal beamformers for passive IRS-assisted sumrate maximization. in ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)., 10095269, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, vol. 2023-June, IEEE, pp. 1-5, ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Greece, 4/06/23. https://doi.org/10.1109/ICASSP49357.2023.10095269

Model-free learning of optimal beamformers for passive IRS-assisted sumrate maximization. / Hashmi, Hassaan; Pougkakiotis, Spyridon; Kalogerias, Dionysis S.
ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP). IEEE, 2023. p. 1-5 10095269 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2023-June).

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

TY - GEN

T1 - Model-free learning of optimal beamformers for passive IRS-assisted sumrate maximization

AU - Hashmi, Hassaan

AU - Pougkakiotis, Spyridon

AU - Kalogerias, Dionysis S.

PY - 2023/6/10

Y1 - 2023/6/10

N2 - Although Intelligent Reflective Surfaces (IRSs) are a cost-effective technology promising high spectral efficiency in future wireless networks, obtaining optimal IRS beamformers is a challenging problem with several practical limitations. Assuming fully-passive, sensing- free IRS operation, we introduce a new data-driven Zeroth-order Stochastic Gradient Ascent (ZoSGA) algorithm for sumrate optimization in an IRS-aided downlink setting. ZoSGA does not require access to channel model or network structure information, and enables learning of optimal long-term IRS beamformers jointly with standard short-term precoding, based only on conventional effective channel state information. Supported by state-of-the-art (SOTA) convergence analysis, detailed simulations confirm that ZoSGA exhibits SOTA empirical behavior as well, consistently outperforming standard fully model-based baselines, in a variety of scenarios.

AB - Although Intelligent Reflective Surfaces (IRSs) are a cost-effective technology promising high spectral efficiency in future wireless networks, obtaining optimal IRS beamformers is a challenging problem with several practical limitations. Assuming fully-passive, sensing- free IRS operation, we introduce a new data-driven Zeroth-order Stochastic Gradient Ascent (ZoSGA) algorithm for sumrate optimization in an IRS-aided downlink setting. ZoSGA does not require access to channel model or network structure information, and enables learning of optimal long-term IRS beamformers jointly with standard short-term precoding, based only on conventional effective channel state information. Supported by state-of-the-art (SOTA) convergence analysis, detailed simulations confirm that ZoSGA exhibits SOTA empirical behavior as well, consistently outperforming standard fully model-based baselines, in a variety of scenarios.

KW - Intelligent Reflecting Surfaces

KW - Model-Free Learning

KW - Sumrate Maximization

KW - Zeroth-order Optimization

UR - http://www.scopus.com/inward/record.url?scp=85177579762&partnerID=8YFLogxK

U2 - 10.1109/ICASSP49357.2023.10095269

DO - 10.1109/ICASSP49357.2023.10095269

M3 - Conference contribution

SN - 9781728163284

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 1

EP - 5

BT - ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

PB - IEEE

T2 - ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Y2 - 4 June 2023 through 10 June 2023

ER -

Hashmi H, Pougkakiotis S, Kalogerias DS. Model-free learning of optimal beamformers for passive IRS-assisted sumrate maximization. In ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP). IEEE. 2023. p. 1-5. 10095269. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). doi: 10.1109/ICASSP49357.2023.10095269

Model-free learning of optimal beamformers for passive IRS-assisted sumrate maximization

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this