Deep Augmented MUSIC Algorithm for Data-Driven DoA Estimation

Julian P. Merkofer; Guy Revach; Nir Shlezinger; Ruud J.G. van Sloun

doi:10.1109/ICASSP43922.2022.9746637

Deep Augmented MUSIC Algorithm for Data-Driven DoA Estimation

Julian P. Merkofer, Guy Revach, Nir Shlezinger, Ruud J.G. van Sloun

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/Congresprocedure › Conferentiebijdrage › Academic › peer review

22 Citaten (Scopus)

1 Downloads (Pure)

Samenvatting

Direction of arrival (DoA) estimation is a crucial task in sensor array signal processing, giving rise to various successful model-based (MB) algorithms as well as recently developed data-driven (DD) methods. This paper introduces a new hybrid MB/DD DoA estimation architecture, based on the classical multiple signal classification (MUSIC) algorithm. Our approach augments crucial aspects of the original MUSIC structure with specifically designed neural architectures, allowing it to overcome certain limitations of the purely MB method, such as its inability to successfully localize coherent sources. The deep augmented MUSIC algorithm is shown to outperform its unaltered version with a superior resolution.

Originele taal-2	Engels
Titel	2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings
Uitgeverij	Institute of Electrical and Electronics Engineers
Pagina's	3598-3602
Aantal pagina's	5
ISBN van elektronische versie	9781665405409
DOI's	https://doi.org/10.1109/ICASSP43922.2022.9746637
Status	Gepubliceerd - 27 apr. 2022
Evenement	ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) - Virtual, Online, Singapore, Singapore Duur: 23 mei 2022 → 27 mei 2022 Congresnummer: 47 https://2022.ieeeicassp.org/

Congres

Congres	ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Verkorte titel	ICASSP 2022
Land/Regio	Singapore
Stad	Singapore
Periode	23/05/22 → 27/05/22
Internet adres	https://2022.ieeeicassp.org/

Toegang tot document

10.1109/ICASSP43922.2022.9746637

Andere bestanden en links

Link naar publicatie in Scopus

Citeer dit

Merkofer, J. P., Revach, G., Shlezinger, N., & van Sloun, R. J. G. (2022). Deep Augmented MUSIC Algorithm for Data-Driven DoA Estimation. In 2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings (blz. 3598-3602). Artikel 9746637 Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ICASSP43922.2022.9746637

@inproceedings{d2fa095a98834989b6cff840d3bd70b3,

title = "Deep Augmented MUSIC Algorithm for Data-Driven DoA Estimation",

abstract = "Direction of arrival (DoA) estimation is a crucial task in sensor array signal processing, giving rise to various successful model-based (MB) algorithms as well as recently developed data-driven (DD) methods. This paper introduces a new hybrid MB/DD DoA estimation architecture, based on the classical multiple signal classification (MUSIC) algorithm. Our approach augments crucial aspects of the original MUSIC structure with specifically designed neural architectures, allowing it to overcome certain limitations of the purely MB method, such as its inability to successfully localize coherent sources. The deep augmented MUSIC algorithm is shown to outperform its unaltered version with a superior resolution.",

keywords = "deep learning, DoA estimation, MUSIC",

author = "Merkofer, {Julian P.} and Guy Revach and Nir Shlezinger and {van Sloun}, {Ruud J.G.}",

year = "2022",

month = apr,

day = "27",

doi = "10.1109/ICASSP43922.2022.9746637",

language = "English",

pages = "3598--3602",

booktitle = "2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

note = "2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022, ICASSP 2022 ; Conference date: 23-05-2022 Through 27-05-2022",

url = "https://2022.ieeeicassp.org/",

}

Merkofer, JP, Revach, G, Shlezinger, N & van Sloun, RJG 2022, Deep Augmented MUSIC Algorithm for Data-Driven DoA Estimation. in 2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings., 9746637, Institute of Electrical and Electronics Engineers, blz. 3598-3602, ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Singapore, Singapore, 23/05/22. https://doi.org/10.1109/ICASSP43922.2022.9746637

Deep Augmented MUSIC Algorithm for Data-Driven DoA Estimation. / Merkofer, Julian P.; Revach, Guy; Shlezinger, Nir et al.
2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings. Institute of Electrical and Electronics Engineers, 2022. blz. 3598-3602 9746637.

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/Congresprocedure › Conferentiebijdrage › Academic › peer review

TY - GEN

T1 - Deep Augmented MUSIC Algorithm for Data-Driven DoA Estimation

AU - Merkofer, Julian P.

AU - Revach, Guy

AU - Shlezinger, Nir

AU - van Sloun, Ruud J.G.

N1 - Conference code: 47

PY - 2022/4/27

Y1 - 2022/4/27

N2 - Direction of arrival (DoA) estimation is a crucial task in sensor array signal processing, giving rise to various successful model-based (MB) algorithms as well as recently developed data-driven (DD) methods. This paper introduces a new hybrid MB/DD DoA estimation architecture, based on the classical multiple signal classification (MUSIC) algorithm. Our approach augments crucial aspects of the original MUSIC structure with specifically designed neural architectures, allowing it to overcome certain limitations of the purely MB method, such as its inability to successfully localize coherent sources. The deep augmented MUSIC algorithm is shown to outperform its unaltered version with a superior resolution.

AB - Direction of arrival (DoA) estimation is a crucial task in sensor array signal processing, giving rise to various successful model-based (MB) algorithms as well as recently developed data-driven (DD) methods. This paper introduces a new hybrid MB/DD DoA estimation architecture, based on the classical multiple signal classification (MUSIC) algorithm. Our approach augments crucial aspects of the original MUSIC structure with specifically designed neural architectures, allowing it to overcome certain limitations of the purely MB method, such as its inability to successfully localize coherent sources. The deep augmented MUSIC algorithm is shown to outperform its unaltered version with a superior resolution.

KW - deep learning

KW - DoA estimation

KW - MUSIC

UR - http://www.scopus.com/inward/record.url?scp=85131259278&partnerID=8YFLogxK

U2 - 10.1109/ICASSP43922.2022.9746637

DO - 10.1109/ICASSP43922.2022.9746637

M3 - Conference contribution

AN - SCOPUS:85131259278

SP - 3598

EP - 3602

BT - 2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings

PB - Institute of Electrical and Electronics Engineers

T2 - 2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022

Y2 - 23 May 2022 through 27 May 2022

ER -

Deep Augmented MUSIC Algorithm for Data-Driven DoA Estimation

Samenvatting

Congres

Toegang tot document

Andere bestanden en links

Vingerafdruk

Citeer dit