Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

The Online Data Filter for the KM3NeT Neutrino Telescopes

O. Adriani, S. Aiello, A. Albert, A. R. Alhebsi, M. Alshamsi, S. Alves Garre, A. Ambrosone, F. Ameli, M. Andre, L. Aphecetche, M. Ardid, S. Ardid, J. Aublin, F. Badaracco, L. Bailly-Salins, Z. Bardacova, B. Baret, A. Bariego-Quintana, Y. Becherini, M. Bendahman, F. Benfenati Gualandi, M. Benhassi, M. Bennani, D. M. Benoit, E. Berbee, E. Berti, V. Bertin, P. Betti, S. Biagi, M. Boettcher, D. Bonanno, S. Bottai, A. B. Bouasla, J. Boumaaza, M. Bouta, M. Bouwhuis, C. Bozza, R. M. Bozza, H. Branzas, F. Bretaudeau, M. Breuhaus, R. Bruijn, J. Brunner, R. Bruno, E. Buis, R. Buompane, J. Busto, B. Caiffi, D. Calvo, A. Capone, F. Carenini, V. Carretero, T. Cartraud, P. Castaldi, V. Cecchini, S. Celli, L. Cerisy, M. Chabab, A. Chen, S. Cherubini, T. Chiarusi, M. Circella, R. Clark, R. Cocimano, J. A. B. Coelho, A. Coleiro, A. Condorelli, R. Coniglione, P. Coyle, A. Creusot, G. Cuttone, R. Dallier, A. De Benedittis, G. De Wasseige, V. Decoene, P. Deguire, I. Del Rosso, L. S. Di Mauro, I. Di Palma, A. F. Diaz, D. Diego-Tortosa, C. Distefano, A. Domi, C. Donzaud, D. Dornic, E. Drakopoulou, D. Drouhin, J. -G. Ducoin, P. Duverne, R. Dvornicky, T. Eberl, E. Eckerova, A. Eddymaoui, T. van Eeden, M. Eff, D. van Eijk, I. El Bojaddaini, S. El Hedri, S. El Mentawi, A. Enzenhofer, G. Ferrara, M. D. Filipovic, F. Filippini, D. Franciotti, L. A. Fusco, T. Gal, J. Garcia Mendez, A. Garcia Soto, C. Gatius Oliver, N. Geiselbrecht, E. Genton, H. Ghaddari, L. Gialanella, B. K. Gibson, E. Giorgio, I. Goos, P. Goswami, S. R. Gozzini, R. Gracia, B. Guillon, C. Haack, H. van Haren, A. Heijboer, L. Hennig, J. J. Hernandez-Rey, A. Idrissi, W. Idrissi Ibnsalih, G. Illuminati, R. Jaimes, O. Janik, D. Joly, M. de Jong, P. de Jong, B. J. Jung, P. Kalaczynski, J. Keegans, V. Kikvadze, G. Kistauri, C. Kopper, A. Kouchner, Y. Y. Kovalev, L. Krupa, V. Kueviakoe, V. Kulikovskiy, R. Kvatadze, M. Labalme, R. Lahmann, M. Lamoureux, G. Larosa, C. Lastoria, J. Lazar, A. Lazo, S. Le Stum, G. Lehaut, V. Lemaitre, E. Leonora, N. Lessing, G. Levi, M. Lindsey Clark, F. Longhitano, F. Magnani, J. Majumdar, L. Malerba, F. Mamedov, A. Manfreda, A. Manousakis, M. Marconi, A. Margiotta, A. Marinelli, C. Markou, L. Martin, M. Mastrodicasa, S. Mastroianni, J. Mauro, K. C. K. Mehta, G. Miele, P. Migliozzi, E. Migneco, M. L. Mitsou, C. M. Mollo, L. Morales-Gallegos, N. Mori, A. Moussa, I. Mozun Mateo, R. Muller, M. R. Musone, M. Musumeci, S. Navas, A. Nayerhoda, C. A. Nicolau, B. Nkosi, B. O Fearraigh, V. Oliviero, A. Orlando, E. Oukacha, L. Pacini, D. Paesani, J. Palacios Gonzalez, G. Papalashvili, P. Papini, V. Parisi, A. Parmar, E. J. Pastor Gomez, C. Pastore, A. M. Paun, G. E. Pavalas, S. Pena Martinez, M. Perrin-Terrin, V. Pestel, M. Petropavlova, P. Piattelli, A. Plavin, C. Poire, T. Pradier, J. Prado, S. Pulvirenti, C. A. Quiroz-Rangel, N. Randazzo, A. Ratnani, S. Razzaque, I. C. Rea, D. Real, G. Riccobene, J. Robinson, A. Romanov, E. Ros, A. Saina, F. Salesa Greus, D. F. E. Samtleben, A. Sanchez Losa, S. Sanfilippo, M. Sanguineti, D. Santonocito, P. Sapienza, M. Scaringella, M. Scarnera, J. Schnabel, J. Schumann, J. Seneca, N. Sennan, P. A. Sevle Myhr, I. Sgura, R. Shanidze, Chengyu Shao, A. Sharma, Y. Shitov, F. Simkovic, A. Simonelli, A. Sinopoulou, B. Spisso, M. Spurio, O. Starodubtsev, D. Stavropoulos, I. Stekl, D. Stocco, M. Taiuti, G. Takadze, Y. Tayalati, H. Thiersen, S. Thoudam, I. Tosta e Melo, B. Trocme, V. Tsourapis, E. Tzamariudaki, A. Ukleja, A. Vacheret, V. Valsecchi, V. Van Elewyck, G. Vannoye, E. Vannuccini, G. Vasileiadis, F. Vazquez de Sola, A. Veutro, S. Viola, D. Vivolo, A. van Vliet, E. de Wolf, I. Lhenry-Yvon, S. Zavatarelli, D. Zito, J. D. Zornoza, J. Zuniga

TL;DR

The KM3NeT online data filter addresses the challenge of sifting vast deep-sea detector data in real time by employing an all-data-to-shore architecture with multi-level triggers (L0-L2, Supernova, and summary streams). It uses time-sliced UDP data, FPGA-based offshore digitization, and a farm of commodity servers on shore to achieve an approximate data reduction of $\sim10^3$ while preserving genuine neutrino events. Key contributions include a causality-based trigger framework (CLIQUE for largest causal subset), muon/shower specific event signatures with geometry-driven windows, and a flexible system that supports external alerts and offline reprocessing without compromising live performance. The results show high purity ($<1\%$) and manageable capacity (<$50 CPU cores per block$) across expected PMT rates, with effective volumes approaching the detector geometry at relevant energies, and a practical path to scalable, low-cost operation of KM3NeT as construction proceeds.

Abstract

The KM3NeT research infrastructure comprises two neutrino telescopes located in the deep waters of the Mediterranean Sea, namely ORCA and ARCA. KM3NeT/ORCA is designed for the measurement of neutrino properties and KM3NeT/ARCA for the detection of high-energy neutrinos from the cosmos. Neutrinos are indirectly detected using three-dimensional arrays of photo-sensors which detect the Cherenkov light that is produced when relativistic charged particles emerge from a neutrino interaction. The analogue pulses from the photo-sensors are digitised offshore and all digital data are sent to a station on shore where they are processed in real time using a farm of commodity servers and custom software. In this paper, the design and performance of the software that is used to filter the data are presented. The performance of the data filter is evaluated in terms of its efficiency, purity and capacity. The efficiency is measured by the effective volumes of the sensor arrays as a function of the energy of the neutrino. The purity is measured by a comparison of the event rate caused by muons produced by cosmic ray interactions in the Earth's atmosphere with the event rate caused by the background from decays of radioactive elements in the sea water and bioluminescence. The capacity is measured by the minimal number of servers that is needed to sustain the rate of incoming data. The results of these evaluations comply with all specifications. The count rates of all photo-sensors are measured with a sampling frequency of 10 Hz. These data are input to the simulations of the detector response and will also be made available for interdisciplinary research.

The Online Data Filter for the KM3NeT Neutrino Telescopes

TL;DR

The KM3NeT online data filter addresses the challenge of sifting vast deep-sea detector data in real time by employing an all-data-to-shore architecture with multi-level triggers (L0-L2, Supernova, and summary streams). It uses time-sliced UDP data, FPGA-based offshore digitization, and a farm of commodity servers on shore to achieve an approximate data reduction of while preserving genuine neutrino events. Key contributions include a causality-based trigger framework (CLIQUE for largest causal subset), muon/shower specific event signatures with geometry-driven windows, and a flexible system that supports external alerts and offline reprocessing without compromising live performance. The results show high purity () and manageable capacity (<) across expected PMT rates, with effective volumes approaching the detector geometry at relevant energies, and a practical path to scalable, low-cost operation of KM3NeT as construction proceeds.

Abstract

The KM3NeT research infrastructure comprises two neutrino telescopes located in the deep waters of the Mediterranean Sea, namely ORCA and ARCA. KM3NeT/ORCA is designed for the measurement of neutrino properties and KM3NeT/ARCA for the detection of high-energy neutrinos from the cosmos. Neutrinos are indirectly detected using three-dimensional arrays of photo-sensors which detect the Cherenkov light that is produced when relativistic charged particles emerge from a neutrino interaction. The analogue pulses from the photo-sensors are digitised offshore and all digital data are sent to a station on shore where they are processed in real time using a farm of commodity servers and custom software. In this paper, the design and performance of the software that is used to filter the data are presented. The performance of the data filter is evaluated in terms of its efficiency, purity and capacity. The efficiency is measured by the effective volumes of the sensor arrays as a function of the energy of the neutrino. The purity is measured by a comparison of the event rate caused by muons produced by cosmic ray interactions in the Earth's atmosphere with the event rate caused by the background from decays of radioactive elements in the sea water and bioluminescence. The capacity is measured by the minimal number of servers that is needed to sustain the rate of incoming data. The results of these evaluations comply with all specifications. The count rates of all photo-sensors are measured with a sampling frequency of 10 Hz. These data are input to the simulations of the detector response and will also be made available for interdisciplinary research.

Paper Structure

This paper contains 21 sections, 2 equations, 4 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Data processing
  3. Offshore data acquisition
  4. Data transmission
  5. On shore data processing
  6. Data recording
  7. Specifications
  8. Data filter
  9. Level-0
  10. Level-1
  11. Level-2
  12. Supernova
  13. Summary
  14. Trigger
  15. Event signatures
  16. ...and 6 more sections

Figures (4)

  • Figure 1: Schematic view of the data flow (not to scale). The part above the dashed line corresponds to offshore and the part below to on shore. The circles correspond to the optical modules which are deployed in the deep sea and house the PMTs as well as the FPGA. The ellipses and rectangular boxes correspond to processes and switch fabric, respectively. The base provides for the connection to the sea-floor network and is mounted on the anchor of the string.
  • Figure 2: The time windows for the KM3NeT/ARCA neutrino telescope as a function of the distance between optical modules for the different causality relations (see text). The left (right) figure applies to the signal from a muon (shower). The maximal distance roughly corresponds to the size of the detector. The label 3D corresponds to the general causality relation, 3B to the causality relation for which the distance covered by the signal is decomposed and 1D to the causality relation for which the propagation time of the muon is taken into account. For the 1D case, the x-axis corresponds to the transverse distance with respect to the direction of the muon. The label 3G corresponds to the causality relation for which the distance covered by the signal is limited to the radius of a sphere.
  • Figure 3: The rate of events (left) and the required number of CPU cores (right) as a function of the singles rate of the PMTs for one building block of the KM3NeT/ARCA and KM3NeT/ORCA neutrino telescopes. The points correspond to the results obtained from a simulation of the random background. The dashed lines correspond to the expected rates of events due to muons produced by interactions of cosmic rays in the atmosphere above the detectors. A single CPU core of the Intel(R) Xeon(R) E E-2478 processor is used.
  • Figure 4: The effective volume of one building block of the KM3NeT/ORCA (top) and KM3NeT/ARCA (bottom) neutrino telescopes as a function of the neutrino energy and averaged over all directions for charged-current interactions of electron (left) and muon (right) neutrinos (note the scale). The labels correspond to the signature that is used in the data filter. The dashed lines correspond to the geometrical volumes of the detectors.