Study On The B-tagging Algorithm In ATLAS

The world largest proton-proton collider--Large Hadron Collider(LHC),which is designed by the European Organization for Nuclear Research(CERN),is supposed to start doing physics in 2008.The Large Hadron Collider(LHC)mainly consists of a 27 km ring of superconducting magnets with a number of accelerating structures about 100 m underground.The high energies reached by the LHC(a centre of mass energy of 14 TeV and a design luminosity of 10³⁴cm^-2s^-1)can be used to validate the Standard Model or search Higgs particle.There are four experiments under construction at the colliding points of LHC: ATLAS(A Toroidal LHC ApparatuS),CMS(Compact Muon Solenoid),ALICE (A Large Ion Collider Experiment)and LHCb.The ATLAS experiment is one of two General Purpose Detectors at LHC.It was designed to exploit the full discovery potential of LHC,to search for Higgs particle and heavy W and Z- objects,for super-symmetric particles,for compositeness of the fundamental fermions,and to investigate the CP violation in B-physics.The ATLAS detector consists of four major components,the Inner Detector,the Calorimeter,the Muon spectrometer,and the Magnet system.For decades,the Standard Model of particle physics has served physicists well as a means of understanding the fundamental laws of nature,but it does not tell the whole story.To date,almost all experimental tests of the three forces described by the standard model have agreed with its predictions.Currently the Standard Model predicts that there is one more particle to be discovered,the Higgs boson.The Large Hadron Collider is expected to find Higgs particle.The importance that ATLAS places on b-tagging for LHC physics has increased over the past year,as a result of recent detailed studies of Higgs and supersymmetry(SUSY)signatures,which were performed for the ATLAS Technical Proposal(TP)and for the discussions with the LHC referees.B-Tagging in ATLAS plays a relevant role in many analyses where the signal events with one or more bottom-quark Jets in the final state have to be separated from a huge amount of background events which mainly contain up-,down-and strange-quark Jets or c-quark Jets.The very large non-b Jet backgrounds often demand that all b-quarks present in the final state under study be identified.Therefore b-tagging with as high an efficiency as possible,using impact parameter algorithms together with the vertexing algorithms,will be one of the keys to a successful search for new physics at LHC.This has been pioneered with great success by the CDF collaboration at the Fermilab Tevatron,for the discovery of the top quark.There are 3 classes of b-tagging algorithms implemented in ATLAS so far:(â…°)Impact parameter significance of tracks(with respect to the reconstructed primary interaction point)(â…±)Reconstruction of a displaced vertex and its properties(â…²)Reconstruction of a soft lepton from the semi-leptonic B decay inside the JetThe default ATLAS b-tagging algorithm makes use of a combination of the first two methods.The purpose of this thesis is to investigate the possibility of separating the b-Jet from large numbers of u-Jet background.In this thesis,several b-tagging algorithms were studied by using MC samples.Using ten variables as inputs,multi-variant analysis is performed with TMVA(Toolkit for Multivariate Analysis)program to separate the b-Jet from u-Jet.