Speaker
Lukas Berns
(Tokyo Institute of Technology)
Description
We have proposed the possibility of a cost-efficient way to improve the detector performance for water Cherenkov detectors, by reflecting the usually lost light falling between photo-detectors onto the other side of the tank with retro-reflectors. Using a detector simulation based on optical measurements of retro-reflectors, we developed a convolutional neural network based reconstruction algorithm. Here we report on the reconstruction performance for ring events in the energy scale expected for atmospheric and accelerator neutrinos under various candidate detector configurations.
Primary author
Lukas Berns
(Tokyo Institute of Technology)
Co-authors
Shunsuke Fujigami
(Tokyo Institute of Technology)
Masahiro Kuze
(Tokyo Institute of Technology)
Yohei Yamaguchi
(Tokyo Institute of Technology)
