Improving sensitivity of the ARIANNA detector by rejecting thermal noise with deep learning

Anker A, Baldi P, Barwick SW, Beise J, Besson DZ, Bouma S, Cataldo M, Chen P, Gaswint G, Glaser C, Hallgren A, Hallmann S, Hanson JC, Klein SR, Kleinfelder SA, Lahmann R, Liu J, Magnuson M, Mcaleer S, Meyers Z, Nam J, Nelles A, Novikov A, Paul MP, Persichilli C, Plaisier I, Pyras LM, Rice-Smith R, Tatar J, Wang SH, Welling C, Zhao L (2022)

Publication Type: Journal article

Publication year: 2022

Journal

Journal of Instrumentation Institute of Physics: Hybrid Open Access

Book Volume: 17

Article Number: P03007

Journal Issue: 3

DOI: 10.1088/1748-0221/17/03/P03007

Abstract

The ARIANNA experiment is an Askaryan detector designed to record radio signals induced by neutrino interactions in the Antarctic ice. Because of the low neutrino flux at high energies (Eν > 1016 eV), the physics output is limited by statistics. Hence, an increase in sensitivity significantly improves the interpretation of data and offers the ability to probe new parameter spaces. The amplitudes of the trigger threshold are limited by the rate of triggering on unavoidable thermal noise fluctuations. We present a real-time thermal noise rejection algorithm that enables the trigger thresholds to be lowered, which increases the sensitivity to neutrinos by up to a factor of two (depending on energy) compared to the current ARIANNA capabilities. A deep learning discriminator, based on a Convolutional Neural Network (CNN), is implemented to identify and remove thermal events in real time. We describe a CNN trained on MC data that runs on the current ARIANNA microcomputer and retains 95% of the neutrino signal at a thermal noise rejection factor of 105, compared to a template matching procedure which reaches only 102 for the same signal efficiency. Then the results are verified in a lab measurement by feeding in generated neutrino-like signal pulses and thermal noise directly into the ARIANNA data acquisition system. Lastly, the same CNN is used to classify cosmic-rays events to make sure they are not rejected. The network classified 102 out of 104 cosmic-ray events as signal.

Authors with CRIS profile

Sjoerd Bouma Professur für Experimentelle Astroteilchenphysik, insbesondere Radionachweis von Neutrinos Maddalena Cataldo Professur für Experimentelle Astroteilchenphysik, insbesondere Radionachweis von Neutrinos Robert Lahmann Lehrstuhl für Experimentalphysik (Astroteilchenphysik) Zachary Meyers Professur für Experimentelle Astroteilchenphysik, insbesondere Radionachweis von Neutrinos Anna Nelles Professur für Experimentelle Astroteilchenphysik, insbesondere Radionachweis von Neutrinos Lilly Marie Pyras Professur für Experimentelle Astroteilchenphysik, insbesondere Radionachweis von Neutrinos Christoph Welling Professur für Experimentelle Astroteilchenphysik, insbesondere Radionachweis von Neutrinos

Involved external institutions

National Taiwan University (NTU) TW Taiwan (TW) University of California Irvine US United States (USA) (US) Uppsala University SE Sweden (SE) University of Kansas (KU) US United States (USA) (US) Deutsches Elektronen-Synchrotron DESY DE Germany (DE) Lawrence Berkeley National Laboratory (LBNL) US United States (USA) (US) Whittier College US United States (USA) (US)

How to cite

APA:

Anker, A., Baldi, P., Barwick, S.W., Beise, J., Besson, D.Z., Bouma, S.,... Zhao, L. (2022). Improving sensitivity of the ARIANNA detector by rejecting thermal noise with deep learning. Journal of Instrumentation, 17(3). https://doi.org/10.1088/1748-0221/17/03/P03007

MLA:

Anker, A., et al. "Improving sensitivity of the ARIANNA detector by rejecting thermal noise with deep learning." Journal of Instrumentation 17.3 (2022).

BibTeX: Download