NEURAL 2ND-LEVEL TRIGGER SYSTEM BASED ON CALORIMETRY

A second-level triggering system based on calorimetry is analyzed usin g neural networks. Calorimeter data in a LHC environment is obtained w ith Monte Carlo simulations and an algorithm for the first-level trigg er operation is applied. The surviving events are then available as a 20x20 matrix information corresponding to the calorimeter towers in th e region of interest. The dominant background for triggering on electr ons is assumed to consist of QCD jets which passed the first-level tri gger condition. The main features of the calorimeter are extracted. Ma trix information, shower deposition in concentric rings and tail weigh ting procedures are studied. The processed information is sent to a fu lly connected backpropagation neural network. In this analysis we also consider pileup effects of an average of 20 minimum bias events. The neural network based system achieved up to 99% electron efficiency wit h less than 9% of jets being misclassified as electrons. Implementatio n on digital signal processors is suggested.