- Physical Review Journals
All JournalsPhysics Magazine

Physical Review D

Reuse & Permissions

It is not necessary to obtain permission to reuse this article or its components as it is available under the terms of the Creative Commons Attribution 4.0 International license. This license permits unrestricted use, distribution, and reproduction in any medium, provided attribution to the author(s) and the published article's title, journal citation, and DOI are maintained. Please note that some figures may have been included with permission from other third parties. It is your responsibility to obtain the proper permission from the rights holder directly for these figures.

Export citation

Export citation

Choose format for download:

Download Citation
  • Open Access

Deep-learning denoising of radio signals for ultra-high-energy cosmic-ray detection

Zhisen Lai1,*, Oscar Macias1,2,†, Aurélien Benoit-Lévy3, Arsène Ferrière3,4, and Matías Tueros5,6

  • *Contact author: zlai@sfsu.edu
  • †Contact author: macias@sfsu.edu
PDF

Phys. Rev. D 114, 062011 – Published 28 September, 2026

DOI: https://doi.org/10.1103/29jx-bcgq

Abstract

Extensive air showers initiated by ultra-high-energy cosmic rays produce broadband radio pulses detectable over large areas, but Galactic and instrumental backgrounds limit sensitivity near threshold. The Giant Radio Array for Neutrino Detection (GRAND) will use large arrays of autonomous radio antennas to detect these air showers, requiring reliable recovery of weak pulses from background fluctuations. We develop a convolutional denoiser for GRAND-like antenna traces that combines the time-domain waveform with the magnitude and phase of its Fourier transform. The model is trained on 4.1×105 simulated traces constructed from zhaires air-shower emission, a reduced GRAND-like radio-frequency (RF) chain response, and broadband noise normalized to the expected Galactic-noise scale. In these simulations, denoising increases the detection probability at fixed false-positive rate and improves pulse-time recovery near threshold. The reconstructed peak amplitude is attenuated in the noise-dominated regime. This bias decreases as the pulse becomes better resolved, but a smaller channel-dependent bias persists at higher signal-to-noise ratio. Near threshold, denoising therefore primarily improves candidate recovery and timing, while the measured amplitude response provides a basis for calibrating amplitude-sensitive observables.

View figure in article

Physics Subject Headings (PhySH)

Article Text

References (38)

  1. L. A. Anchordoqui, Ultra-high-energy cosmic rays, Phys. Rep. 801, 1 (2019).
  2. R. Alves Batista, The quest for the origins of ultra-high-energy cosmic rays, in 28th European Cosmic Ray Symposium (2024), arXiv:2412.17201.
  3. J. Álvarez-Muñiz et al. (GRAND Collaboration), The Giant Radio Array for Neutrino Detection (GRAND): Science and design, Sci. China Phys. Mech. Astron. 63, 219501 (2020).
  4. T. Huege, Radio detection of cosmic ray air showers in the digital era, Phys. Rep. 620, 1 (2016).
  5. F. G. Schröder, Radio detection of cosmic-ray air showers and high-energy neutrinos, Prog. Part. Nucl. Phys. 93, 1 (2017).
  6. D. Shipilov et al., Signal recognition and background suppression by matched filters and neural networks for Tunka-Rex, EPJ Web Conf. 216, 02003 (2019).
  7. C. K. O. Watanabe, P. S. R. Diniz, T. Huege, and J. de Mello Neto, Denoising cosmic rays radio signal using wavelets techniques, Proc. Sci. ICRC2021 (2021) 260.
  8. J. A. Högbom, Aperture synthesis with a non-regular distribution of interferometer baselines, Astron. Astrophys. Suppl. Ser. 15, 417 (1974), https://ui.adsabs.harvard.edu/abs/1974A%26AS...15..417H/abstract.
  9. A. Horneffer et al. (LOPES Collaboration), LOPES: Detecting radio emission from cosmic ray air showers, Proc. SPIE Int. Soc. Opt. Eng. 5500, 129 (2004).
  10. S. Martinelli, T. Huege, D. Ravignani, and H. Schoorlemmer, Quantifying energy fluence and its uncertainty for radio emission from particle cascades in the presence of noise, Astropart. Phys. 168, 103091 (2025).
  11. C. Gheller and F. Vazza, Convolutional deep denoising autoencoders for radio astronomical images, Mon. Not. R. Astron. Soc. 509, 990–1009 (2022).
  12. M. Terris, A. Dabbech, C. Tang, and Y. Wiaux, Image reconstruction algorithms in radio interferometry: From handcrafted to learned regularization denoisers, Mon. Not. R. Astron. Soc. 518, 604–622 (2023).
  13. L. Einig et al., Deep learning denoising by dimension reduction: Application to the ORION-B line cubes, Astron. Astrophys. 677, A158 (2023).
  14. C. Reissel, S. Soni, M. Saleem, M. Coughlin, P. Harris, and E. Katsavounidis, Toward autonomous denoising of gravitational-wave detector data, Phys. Rev. D 113, 122007 (2026).
  15. M. Erdmann, F. Schlüter, and R. Smida, Classification and recovery of radio signals from cosmic ray induced air showers with deep learning, J. Instrum. 14, P04005 (2019).
  16. P. Bezyazeekov, D. Shipilov, D. Kostunin, I. Plokhikh, A. Mikhaylenko, P. Turishcheva, S. Golovachev, V. Sotnikov, and E. Sotnikova, Reconstruction of sub-threshold events of cosmic-ray radio detectors using an autoencoder, in Proceedings of 37th International Cosmic Ray Conference—PoS(ICRC2021), ICRC2021 (, Trieste, Italy, 2021), p. 223.
  17. F. G. Schroeder and A. Rehman (IceCube Collaboration), Application of machine learning to identify radio pulses of air showers at the South Pole (ARENA 2024), Proc. Sci. ARENA2024 (2024) 034. [arXiv:2502.19931].
  18. R. Abbasi et al. (IceCube Collaboration), Identification and denoising of radio signals from cosmic-ray air showers using convolutional neural networks, Phys. Rev. D 113, 122002 (2026).
  19. A. Benoit-Levy, Z. Lai, O. Macias, and A. Ferriere (GRAND Collaboration), Denoising radio pulses from air showers using machine-learning methods, Proc. Sci. ICRC2025 (2025) 185.
  20. P. Correa and J. Köhler (GRAND Collaboration), NUTRIG: Development of a novel radio self-trigger for GRAND, Proc. Sci. ICRC2025 (2025) 229. [arXiv:2507.04339].
  21. S. Chiche, GRANDProto300: Status, science case, and prospects, arXiv:2409.02195.
  22. J. Alvarez-Muñiz, W. R. Carvalho, M. Tueros, and E. Zas, Coherent Cherenkov radio pulses from hadronic showers up to EeV energies, Astropart. Phys. 35, 287 (2012).
  23. C. S. C. Sanchez, P. M. Hansen, M. Tueros, J. Alvarez-Muñiz, and D. G. Melo, Uncertainties in the estimation of air shower observables from Monte Carlo simulation of radio emission, Astropart. Phys. 175, 103183 (2026).
  24. M. Guelfand (GRAND Collaboration), The Giant Radio Array for Neutrino Detection (GRAND) and its prototype phases, in 58th Rencontres de Moriond on Very High Energy Phenomena in the Universe (2025), arXiv:2501.01851.
  25. R. Alves Batista et al. (GRAND Collaboration), GRANDlib: A simulation pipeline for the Giant Radio Array for Neutrino Detection (GRAND), Comput. Phys. Commun. 308, 109461 (2025).
  26. P. Vincent, H. Larochelle, Y. Bengio, and P.-A. Manzagol, Extracting and composing robust features with denoising autoencoders, in Proceedings of the 25th International Conference on Machine Learning, ACM International Conference Proceeding Series Vol. 307 (, New York, 2008), pp. 1096–1103, 10.1145/1390156.1390294.
  27. K. He, X. Zhang, S. Ren, and J. Sun, Identity mappings in deep residual networks, arXiv:1603.05027.
  28. Z. Lai, O. Macias, and A. Benoit-Lévy, Deep-learning denoising of radio observations for ultra-high-energy cosmic-ray detection, Zenodo (2026), software, raytune_lib_final release, 10.5281/zenodo.18233878.
  29. Z. Lai, O. Macias, A. Benoit-Lévy, A. Ferrière, and M. Tueros, Radio signal denoising with CNN and ray tune, GitHub (2026), software and analysis products, raytune_lib_final_v1 branch (accessed September 10, 2026), https://github.com/grand-mother/ML_denoising, https://github.com/grand-mother/ML_denoising/tree/raytune_lib_final_v1.
  30. A. Paszke, S. Gross, F. Massa, A. Lerer, J. Bradbury, G. Chanan, T. Killeen, Z. Lin, N. Gimelshein, L. Antiga, A. Desmaison, A. Köpf, E. Yang, Z. DeVito, M. Raison, A. Tejani, S. Chilamkurthy, B. Steiner, L. Fang, J. Bai, and S. Chintala, pytorch: An imperative style, high-performance deep learning library, arXiv:1912.01703.
  31. D. P. Kingma and J. Ba, Adam: A method for stochastic optimization, arXiv:1412.6980.
  32. L. N. Smith, Cyclical learning rates for training neural networks, arXiv:1506.01186.
  33. R. Liaw, E. Liang, R. Nishihara, P. Moritz, J. E. Gonzalez, and I. Stoica, Tune: A research platform for distributed model selection and training, arXiv:1807.05118.
  34. L. Li, K. Jamieson, A. Rostamizadeh, E. Gonina, M. Hardt, B. Recht, and A. Talwalkar, A system for massively parallel hyperparameter tuning, arXiv:1810.05934.
  35. A. Ferrière, S. Prunet, A. Benoit-Lévy, M. Guelfand, K. Kotera, and M. Tueros, Analytical planar wavefront reconstruction and error estimates for radio detection of extensive air showers, Nucl. Instrum. Methods Phys. Res., Sect. A 1072, 170178 (2025).
  36. V. Decoene, O. Martineau-Huynh, M. Tueros, and S. Chiche, A reconstruction procedure for very inclined extensive air showers based on radio signals, Proc. Sci. ICRC2021 (2021) 211. [arXiv:2107.03206].
  37. O. Macias, Z. Mason, M. Ho, A. Ferrière, A. Benoit-Lévy, and M. Tueros, Simulation-based inference for direction reconstruction of ultra-high-energy cosmic rays with radio arrays, Phys. Rev. D 113, 063018 (2026).
  38. H. L. Van Trees and K. L. Bell, Detection, Estimation, and Modulation Theory, Part I: Detection, Estimation, and Filtering Theory, 2nd ed. (Wiley, Hoboken, NJ, 2013).

Outline

Information

Phys. Rev. D 114, 062011– Published 28 September, 2026
  • Received 3 February 2026
  • Accepted 8 September 2026
Crossmark - Check for updates button
Creative Commons Logo

Published by the American Physical Society under the terms of the Creative Commons Attribution 4.0 International license. Further distribution of this work must maintain attribution to the author(s) and the published article’s title, journal citation, and DOI. Funded by SCOAP3.

Published by the American Physical Society

Sign In to Your Journals Account

Filter

Filter

Article Lookup

Enter a citation