Identi cation of the hadronic interaction layer in a highly granular calorimeter bymultivariate techniques

Abstract

Calorimetry in future and upgrade collider detectors has a clear overall trend towards high granularity, bothlaterally and longitudinally. This trend is a requirement for the full exploitation of particle ow algorithms, whichreconstruct individual particles using the subdetector that provides the best resolution for this speci c particle. Theincreased level of detail in the event topologies due to higher segmentation of the calorimeter provides additional handlesfor several calorimetric measurements, such as the identi cation of the hadronic interaction layer, which is not possiblefor calorimeters with traditional tower geometry. Here, the power of using multivariate statistical techniques in theidenti cation of the hadronic interaction layer in a highly segmented calorimeter is investigated.