Classification and recovery of radio signals from cosmic ray induced air showers with deep learning

  title={Classification and recovery of radio signals from cosmic ray induced air showers with deep learning},
  author={Martin Erdmann and F. Schlueter and Rabiaa Smida},
  journal={Journal of Instrumentation},
  pages={P04005 - P04005}
Radio emission from air showers enables measurements of cosmic particle kinematics and identity. The radio signals are detected in broadband Megahertz antennas among continuous background noise. We present two deep learning concepts and their performance when applied to simulated data. The first network classifies time traces as signal or background. We achieve a true positive rate of about 90% for signal-to-noise ratios larger than three with a false positive rate below 0.2%. The other network… 
Advanced Signal Reconstruction in Tunka-Rex with Matched Filtering and Deep Learning
The present work shows the comparison between the signal reconstructions obtained with matched filtering and deep neural networks with autoencoder architecture, and gives an example of the first reconstruction of the Tunka-Rex signals obtained with aDeep neural networks.
Deep learning to classify ultra-high-energy cosmic rays by means of PMT signals
This work presents a supervised learning algorithm to determine that the particle is a photon or a hadron, and the combination of automatically extracted features and external variables is able to predict the particle that has produced the air shower with an accuracy of 98.87%.
Digital Radio Arrays for the Detection of Air Showers Initiated by Ultra-High-Energy Particles (Epiphany 2022)
Digital radio arrays have become an effective tool to measure air showers at energies around and above 100 PeV. Compared to optical techniques, the radio technique is not restricted to clear nights.
The Giant Radio Array for Neutrino Detection (GRAND) project
  • K. Kotera
  • Physics
    Proceedings of 37th International Cosmic Ray Conference — PoS(ICRC2021)
  • 2021
TheGRANDproject aims to detect ultra-high-energy neutrinos, cosmic rays and gamma rays, with an array of 200, 000 radio antennas over 200, 000 km2, split into ∼ 20 sub-arrays of ∼ 10, 000 km2
Anomaly Detection in Astrophysics: A Comparison Between Unsupervised Deep and Machine Learning on KiDS Data
Two different unsupervised machine learning algorithms are applied to the problem of anomaly detection, considered as potentially promising methods to detect peculiar sources, a Disentangled Convolutional Autoencoder and an Unsupervised Random Forest.
Data-driven detector signal characterization with constrained bottleneck autoencoders
This article presents how deep learning in the form of constrained bottleneck autoencoders can be used to learn the underlying unknown detector response model directly from data using a waveform toy model.
High-Energy Galactic Cosmic Rays (Astro2020 Science White Paper)
The origin of the highest energy Galactic cosmic rays is still not understood, nor is the transition to EeV extragalactic particles. Scientific progress requires enhancements of existing air-shower
Towards the Tunka-Rex Virtual Observatory
This paper gives a first detailed overview of the concept of the Tunka-Rex Virtual Observatory (TRVO), a framework for open access to the Tunker-Rex data, which currently is under active development and testing.


Radio detection of cosmic ray air showers in the digital era
Towards online triggering for the radio detection of air showers using deep neural networks
An approach to trigger air-shower events on a single-antenna level as well as performing an online reconstruction of the shower parameters using neural networks are described.
Detection and imaging of atmospheric radio flashes from cosmic ray air showers
The results show that it should be possible to determine the nature and composition of UHECRs with combined radio and particle detectors, and to detect the ultrahigh-energy neutrinos expected from flavour mixing.
Signal recognition and background suppression by matched filters and neural networks for Tunka-Rex
The improvements of the signal reconstruction applied for Tunka-Rex are discussed, the comparison between the standard method of signal reconstruction, matched filtering and the autoencoder, and the prospects of application of neural networks for lowering the threshold of digital antenna arrays for cosmic-ray detection are discussed.
Radio detection of Cosmic-Ray Air Showers and High-Energy Neutrinos
Reconstruction of the energy and depth of maximum of cosmic-ray air showers from LOPES radio measurements
LOPES is a digital radio interferometer located at Karlsruhe Institute of Technology (KIT), Germany, which measures radio emission from extensive air showers at MHz frequencies in coincidence with
First Experimental Characterization of Microwave Emission from Cosmic Ray Air Showers.
The measurements show that microwave radiation offers a new means of studying air showers at E≥10^{17}  eV, and are consistent with a mainly forward-directed and polarized emission process in the GHz frequency range.
Energy estimation of cosmic rays with the Engineering Radio Array of the Pierre Auger Observatory
The Auger Engineering Radio Array (AERA) is part of the Pierre Auger Observatory and is used to detect the radio emission of cosmic-ray air showers. These observations are compared to the data of the