A study of tau identification with the CMS detector at the LHC

Abstract

In this thesis I explore the identification of [tau] leptons from simulated reconstructed data that will be collected by the Compact Muon Solenoid (CMS) detector on the Large Hadron Collider (LHC) at CERN. The two components of particle identification, efficiencies of [tau] identification from generator level information, along with fake rates of the current default algorithm have been determined and analyzed for a photon plus jets background sample and QCD background sample. I propose a new [tau] lepton identification algorithm that employs a signal cone parametrized with respect to the 7 transverse energy, and an isolation cone parametrized with respect to charged particle density surrounding the [tau] jet. Using the default algorithm an efficiency of 27.7% is achieved along with a photon plus jets fake rate of 1.96%. Using the proposed algorithm and matching the efficiency of the default algorithm, an efficiency of 26.9% and a fake rate of 0.44% is achieved. Approximately matching fake rates, an efficiency of 37.4% is achieved with a fake rate of 2.36%.