Panes, B., Eckner, C., Hendriks, L., Caron, S., Dijkstra, K., Johannesson, G., et al. (2021). Identification of point sources in gamma rays using U-shaped convolutional neural networks and a data challenge. Astron. Astrophys., 656, A62–18pp.
Abstract: Context. At GeV energies, the sky is dominated by the interstellar emission from the Galaxy. With limited statistics and spatial resolution, accurately separating point sources is therefore challenging. Aims. Here we present the first application of deep learning based algorithms to automatically detect and classify point sources from gamma-ray data. For concreteness we refer to this approach as AutoSourceID. Methods. To detect point sources, we utilized U-shaped convolutional networks for image segmentation and k-means for source clustering and localization. We also explored the Centroid-Net algorithm, which is designed to find and count objects. Using two algorithms allows for a cross check of the results, while a combination of their results can be used to improve performance. The training data are based on 9.5 years of exposure from The Fermi Large Area Telescope (Fermi-LAT) and we used source properties of active galactic nuclei (AGNs) and pulsars (PSRs) from the fourth Fermi-LAT source catalog in addition to several models of background interstellar emission. The results of the localization algorithm are fed into a classification neural network that is trained to separate the three general source classes (AGNs, PSRs, and FAKE sources). Results. We compared our localization algorithms qualitatively with traditional methods and find them to have similar detection thresholds. We also demonstrate the robustness of our source localization algorithms to modifications in the interstellar emission models, which presents a clear advantage over traditional methods. The classification network is able to discriminate between the three classes with typical accuracy of similar to 70%, as long as balanced data sets are used in classification training. We published online our training data sets and analysis scripts and invite the community to join the data challenge aimed to improve the localization and classification of gamma-ray point sources.
|
KM3NeT Collaboration(Aiello, S. et al), Alves Garre, S., Calvo, D., Carretero, V., Garcia Soto, A., Gozzini, S. R., et al. (2023). KM3NeT broadcast optical data transport system. J. Instrum., 18(2), T02001–22pp.
Abstract: The optical data transport system of the KM3NeT neutrino telescope at the bottom of the Mediterranean Sea will provide more than 6000 optical modules in the detector arrays with a point-to-point optical connection to the control stations onshore. The ARCA and ORCA detectors of KM3NeT are being installed at a depth of about 3500 m and 2500 m, respectively and their distance to the control stations is about 100 kilometers and 40 kilometers. In particular, the two detectors are optimised for the detection of cosmic neutrinos with energies above about 1 TeV (ARCA) and for the detection of atmospheric neutrinos with energies in the range 1 GeV-1 TeV (ORCA). The expected maximum data rate is 200 Mbps per optical module. The implemented optical data transport system matches the layouts of the networks of electro-optical cables and junction boxes in the deep sea. For efficient use of the fibres in the system the technology of Dense Wavelength Division Multiplexing is applied. The performance of the optical system in terms of measured bit error rates, optical budget are presented. The next steps in the implementation of the system are also discussed.
|
KM3NeT Collaboration(Aiello, S. et al), Alves Garre, S., Calvo, D., Carretero, V., Colomer, M., Garcia Soto, A., et al. (2022). The KM3NeT multi-PMT optical module. J. Instrum., 17(7), P07038–28pp.
Abstract: The optical module of the KM3NeT neutrino telescope is an innovative multi-faceted large area photodetection module. It contains 31 three-inch photomultiplier tubes in a single 0.44 m diameter pressure-resistant glass sphere. The module is a sensory device also comprising calibration instruments and electronics for power, readout and data acquisition. It is capped with a breakout-box with electronics for connection to an electro-optical cable for power and long-distance communication to the onshore control station. The design of the module was qualified for the first time in the deep sea in 2013. Since then, the technology has been further improved to meet requirements of scalability, cost-effectiveness and high reliability. The module features a sub-nanosecond timing accuracy and a dynamic range allowing the measurement of a single photon up to a cascade of thousands of photons, suited for the measurement of the Cherenkov radiation induced in water by secondary particles from interactions of neutrinos with energies in the range of GeV to PeV. A distributed production model has been implemented for the delivery of more than 6000 modules in the coming few years with an average production rate of more than 100 modules per month. In this paper a review is presented of the design of the multi-PMT KM3NeT optical module with a proven effective background suppression and signal recognition and sensitivity to the incoming direction of photons.
|
KM3NeT Collaboration(Aiello, S. et al), Alves Garre, S., Calvo, D., Carretero, V., Colomer, M., Corredoira, I., et al. (2020). Event reconstruction for KM3NeT/ORCA using convolutional neural networks. J. Instrum., 15(10), P10005–39pp.
Abstract: The KM3NeT research infrastructure is currently under construction at two locations in the Mediterranean Sea. The KM3NeT/ORCA water-Cherenkov neutrino detector off the French coast will instrument several megatons of seawater with photosensors. Its main objective is the determination of the neutrino mass ordering. This work aims at demonstrating the general applicability of deep convolutional neural networks to neutrino telescopes, using simulated datasets for the KM3NeT/ORCA detector as an example. To this end, the networks are employed to achieve reconstruction and classification tasks that constitute an alternative to the analysis pipeline presented for KM3NeT/ORCA in the KM3NeT Letter of Intent. They are used to infer event reconstruction estimates for the energy, the direction, and the interaction point of incident neutrinos. The spatial distribution of Cherenkov light generated by charged particles induced in neutrino interactions is classified as shower- or track-like, and the main background processes associated with the detection of atmospheric neutrinos are recognized. Performance comparisons to machine-learning classification and maximum-likelihood reconstruction algorithms previously developed for KM3NeT/ORCA are provided. It is shown that this application of deep convolutional neural networks to simulated datasets for a large-volume neutrino telescope yields competitive reconstruction results and performance improvements with respect to classical approaches.
|
KM3NeT Collaboration(Aiello, S. et al), Barrios-Marti, J., Calvo, D., Coleiro, A., Colomer, M., Gozzini, S. R., et al. (2018). Characterisation of the Hamamatsu photomultipliers for the KM3NeT Neutrino Telescope. J. Instrum., 13, P05035–17pp.
Abstract: The Hamamatsu R12199-023-inch photomultiplier tube is the photodetector chosen for the first phase of the KM3NeT neutrino telescope. About 7000 photomultipliers have been characterised for dark count rate, timing spread and spurious pulses. The quantum efficiency, the gain and the peak-to-valley ratio have also been measured for a sub-sample in order to determine parameter values needed as input to numerical simulations of the detector.
|
