Title :
Artificial neural network as a FPGA trigger for a detection of very inclined air showers
Author :
Szadkowski, Zbigniew ; Pytel, K.
Author_Institution :
Dept. of Phys. & Appl. Inf., Univ. of Lodz, Lodz, Poland
Abstract :
The observation of ultra-high energy neutrinos (UHEνs) has become a priority in experimental astroparticle physics. Neutrinos can interact in the atmosphere (downward-going ν) or in the Earth crust (Earth-skimming ν), producing air showers that can be observed with arrays of detectors at the ground. The surface detector array of the Pierre Auger Observatory can detect these types of cascades. The distinguishing signature for neutrino events is the presence of very inclined showers produced close to the ground (i.e., after having traversed a large amount of atmosphere). Up to now, the Pierre Auger Observatory did not find any candidate on a neutrino event. This imposes competitive limits to the diffuse flux of UHEνs. A very low rate of events potentially generated by neutrinos is a significant challenge for a detection technique and requires both sophisticated algorithms and high-resolution hardware. We present a trigger based on a pipeline artificial neural network implemented in a large FPGA which after learning can recognize traces corresponding to special types of events. The structure of an artificial neural network (ANN) algorithm being developed on the MATLAB platform has been implemented into the fast logic of the biggest FPGA from the Cyclone V E family used for the prototype of the Front-End Board (FEB) for the Auger-Beyond-2015. Several algorithms were tested, however, the Levenberg-Marquardt one (trainlm) seems to be the most efficient. The network was taught: a) to recognize ”old” showers (learning on a basis of real Auger very inclined showers (positive markers) and real standard showers especially triggered by Time over Threshold (negative marker)), b) to recognize ”young” showers (on the basis of simulated ”young” events (positive markers) and standard Auger events as a negative reference). A three-layer neural network being taught with real very inclined Auger showers sh- ws a good efficiency in pattern recognition of 16-point traces with profiles characteristic for ”old” showers. Nevertheless, preliminary simulations of showers with CORSIKA and the response of the water Cherenkov tanks with OffLine suggest that for muonic showers starting a development deeply in the atmosphere ADC traces are not too long and 16-point analysis should be sufficient for a recognition of ”young” showers.
Keywords :
cosmic ray apparatus; cosmic ray muons; cosmic ray neutrinos; cosmic ray showers; field programmable gate arrays; neural nets; trigger circuits; ANN algorithm; Auger-Beyond-2015; CORSIKA; Cyclone V E family; Earth crust; Earth-skimming ν; FEB; FPGA trigger; Front-End Board; Levenberg-Marquardt algorithm; MATLAB platform; OffLine; Pierre Auger Observatory; UHEνs; detection technique; downward-going ν; experimental astroparticle physics; high-resolution hardware; large FPGA; muonic showers; neutrino event; pattern recognition; pipeline artificial neural network; surface detector array; three-layer neural network; ultrahigh energy neutrinos; very inclined air showers detection; very inclined showers; water Cherenkov tanks; young showers; Collaboration; Cosmic rays; Detectors; Field programmable gate arrays; Neutrino sources; Observatories; Protons; DCT; FPGA; MATLAB; Pierre Auger Observatory; neural network; trigger;
Conference_Titel :
Real Time Conference (RT), 2014 19th IEEE-NPSS
Print_ISBN :
978-1-4799-3658-8
DOI :
10.1109/RTC.2014.7097419