Machine learning driven reconstruction of cosmic-ray air showers for next generation radio arrays

IceCube-Gen2 Collaboration

doi:10.22323/1.501.0309

Machine learning driven reconstruction of cosmic-ray air showers for next generation radio arrays

IceCube-Gen2 Collaboration

Physics and Astronomy

Loyola University Chicago
German Electron Synchrotron
University of Canterbury
University of Wisconsin-Madison
Institute of Physics Bhubaneswar
Université libre de Bruxelles
University of Copenhagen
TU Dortmund University
University of Kansas
University of Delaware
Marquette University
Friedrich-Alexander University Erlangen-Nürnberg
Harvard University
University of Utah
RWTH Aachen University
Michigan State University
South Dakota School of Mines & Technology
University of California at Irvine
University of California at Berkeley
Ruhr University Bochum
Chalmers University of Technology
Uppsala University
Technical University of Munich
University of Rochester
University of Maryland, College Park
University of Padua
National Institute for Nuclear Physics
University of Alabama
Karlsruhe Institute of Technology
Johannes Gutenberg University Mainz
Georgia Institute of Technology
Queen's University Kingston
Adelaide University

Research output: Contribution to journal › Conference article › peer-review

Abstract

Surface radio antenna-based measurements of cosmic-ray air showers present significant computational challenges in accurately reconstructing physics observables, in particular, the depth of shower maximum, X_max. State-of-the-art template fitting methods rely on extensive simulation libraries, limiting scalability. This work introduces a technique utilizing graph neural networks to reconstruct key air-shower parameters, in particular, direction and shower-core, energy, and X_max. For training and testing of the networks, we use a CoREAS simulation library made for a future enhancement of IceCube’s surface array with radio antennas. The neural networks provide a scalable framework for large-scale data analysis for next-generation astroparticle observatories, such as IceCube-Gen2.

Original language	English
Article number	309
Journal	Proceedings of Science
Volume	501
DOIs	https://doi.org/10.22323/1.501.0309
State	Published - Dec 30 2025
Event	39th International Cosmic Ray Conference, ICRC 2025 - Geneva, Switzerland Duration: Jul 15 2025 → Jul 24 2025

Access to Document

10.22323/1.501.0309

Cite this

@article{d3119093eeec43a9811a3fc0deac400a,

title = "Machine learning driven reconstruction of cosmic-ray air showers for next generation radio arrays",

abstract = "Surface radio antenna-based measurements of cosmic-ray air showers present significant computational challenges in accurately reconstructing physics observables, in particular, the depth of shower maximum, Xmax. State-of-the-art template fitting methods rely on extensive simulation libraries, limiting scalability. This work introduces a technique utilizing graph neural networks to reconstruct key air-shower parameters, in particular, direction and shower-core, energy, and Xmax. For training and testing of the networks, we use a CoREAS simulation library made for a future enhancement of IceCube{\textquoteright}s surface array with radio antennas. The neural networks provide a scalable framework for large-scale data analysis for next-generation astroparticle observatories, such as IceCube-Gen2.",

author = "\{IceCube-Gen2 Collaboration\} and R. Abbasi and M. Ackermann and J. Adams and Agarwalla, \{S. K.\} and Aguilar, \{J. A.\} and M. Ahlers and Alameddine, \{J. M.\} and S. Ali and Amin, \{N. M.\} and K. Andeen and G. Anton and C. Arg{\"u}elles and Y. Ashida and S. Athanasiadou and J. Audehm and Axani, \{S. N.\} and R. Babu and X. Bai and \{Balagopal V.\}, A. and M. Baricevic and Barwick, \{S. W.\} and V. Basu and R. Bay and \{Becker Tjus\}, J. and P. Behrens and J. Beise and C. Bellenghi and B. Benkel and S. BenZvi and D. Berley and E. Bernardini and Besson, \{D. Z.\} and A. Bishop and E. Blaufuss and L. Bloom and S. Blot and M. Bohmer and F. Bontempo and \{Book Motzkin\}, \{J. Y.\} and J. Borowka and \{Boscolo Meneguolo\}, C. and S. B{\"o}ser and O. Botner and J. B{\"o}ttcher and S. Bouma and J. Braun and B. Brinson and Z. Brisson-Tsavoussis and Burley, \{R. T.\} and J. Kiryluk",

note = "Publisher Copyright: {\textcopyright} Copyright owned by the author(s); 39th International Cosmic Ray Conference, ICRC 2025 ; Conference date: 15-07-2025 Through 24-07-2025",

year = "2025",

month = dec,

day = "30",

doi = "10.22323/1.501.0309",

language = "English",

volume = "501",

journal = "Proceedings of Science",

issn = "1824-8039",

}

TY - JOUR

T1 - Machine learning driven reconstruction of cosmic-ray air showers for next generation radio arrays

AU - IceCube-Gen2 Collaboration

AU - Abbasi, R.

AU - Ackermann, M.

AU - Adams, J.

AU - Agarwalla, S. K.

AU - Aguilar, J. A.

AU - Ahlers, M.

AU - Alameddine, J. M.

AU - Ali, S.

AU - Amin, N. M.

AU - Andeen, K.

AU - Anton, G.

AU - Argüelles, C.

AU - Ashida, Y.

AU - Athanasiadou, S.

AU - Audehm, J.

AU - Axani, S. N.

AU - Babu, R.

AU - Bai, X.

AU - Balagopal V., A.

AU - Baricevic, M.

AU - Barwick, S. W.

AU - Basu, V.

AU - Bay, R.

AU - Becker Tjus, J.

AU - Behrens, P.

AU - Beise, J.

AU - Bellenghi, C.

AU - Benkel, B.

AU - BenZvi, S.

AU - Berley, D.

AU - Bernardini, E.

AU - Besson, D. Z.

AU - Bishop, A.

AU - Blaufuss, E.

AU - Bloom, L.

AU - Blot, S.

AU - Bohmer, M.

AU - Bontempo, F.

AU - Book Motzkin, J. Y.

AU - Borowka, J.

AU - Boscolo Meneguolo, C.

AU - Böser, S.

AU - Botner, O.

AU - Böttcher, J.

AU - Bouma, S.

AU - Braun, J.

AU - Brinson, B.

AU - Brisson-Tsavoussis, Z.

AU - Burley, R. T.

AU - Kiryluk, J.

PY - 2025/12/30

Y1 - 2025/12/30

N2 - Surface radio antenna-based measurements of cosmic-ray air showers present significant computational challenges in accurately reconstructing physics observables, in particular, the depth of shower maximum, Xmax. State-of-the-art template fitting methods rely on extensive simulation libraries, limiting scalability. This work introduces a technique utilizing graph neural networks to reconstruct key air-shower parameters, in particular, direction and shower-core, energy, and Xmax. For training and testing of the networks, we use a CoREAS simulation library made for a future enhancement of IceCube’s surface array with radio antennas. The neural networks provide a scalable framework for large-scale data analysis for next-generation astroparticle observatories, such as IceCube-Gen2.

AB - Surface radio antenna-based measurements of cosmic-ray air showers present significant computational challenges in accurately reconstructing physics observables, in particular, the depth of shower maximum, Xmax. State-of-the-art template fitting methods rely on extensive simulation libraries, limiting scalability. This work introduces a technique utilizing graph neural networks to reconstruct key air-shower parameters, in particular, direction and shower-core, energy, and Xmax. For training and testing of the networks, we use a CoREAS simulation library made for a future enhancement of IceCube’s surface array with radio antennas. The neural networks provide a scalable framework for large-scale data analysis for next-generation astroparticle observatories, such as IceCube-Gen2.

UR - https://www.scopus.com/pages/publications/105029056983

U2 - 10.22323/1.501.0309

DO - 10.22323/1.501.0309

M3 - Conference article

AN - SCOPUS:105029056983

SN - 1824-8039

VL - 501

JO - Proceedings of Science

JF - Proceedings of Science

M1 - 309

T2 - 39th International Cosmic Ray Conference, ICRC 2025

Y2 - 15 July 2025 through 24 July 2025

ER -

Machine learning driven reconstruction of cosmic-ray air showers for next generation radio arrays

Abstract

Access to Document

Other files and links

Fingerprint

Cite this