A feed forward neural network for recognition of fluctuations in electromagnetic showers
Thomas F. Degener · 1993
In order to correlate energy deposits in the electromagnetic calorimeter of the Crystal Barrel detector with particles, neighbouring crystals with energy deposits are grouped to clusters. These clusters are searched for local maxima. Then particles from the primary reaction are assigned to these local maxima. A few percent of these maxima are just due to fluctuations in the electromagnetic cascade. A feed forward neural network is used to separate these fluctuations from energy deposits of photons. 1. Introduction The Crystal Barrel detector has been in operation since October 1989 at the Low Energy Antiproton Ring (LEAR) at CERN. The detector system consists of an electromagnetic calorimeter of 1380 CsI(Tl) crystals surrounding a cylindrical Jet-driftchamber. A more detailed description is given in reference 1 . The experiment is designed to study exclusive charged and neutral final states in antiproton (p) proton (p) annihilation at pbeam momenta up to 1940 MeV/c. This contributio...