DPG Phi
Verhandlungen
Verhandlungen
DPG

Bonn 2010 – scientific programme

Parts | Days | Selection | Search | Downloads | Help

T: Fachverband Teilchenphysik

T 38: Top-Quarks IV

T 38.8: Talk

Thursday, March 18, 2010, 18:30–18:45, HG Aula

Data-Driven Lepton Trigger Efficiencies for Top Decays — •Valentina Ferrara — Humboldt University, Berlin, Germany

One of the first measurements of interest that is expected to be performed with the data collected by ATLAS at the LHC is the production cross-section of tt pairs in pp collisions. According to the Standard Model, each of the two tops decays into a W boson and a b quark. The semi-leptonic channel, namely when only one of the two W bosons decays in a lepton and a neutrino, is a good compromise between high statistics and an acceptable level of background to measure the tt production cross-section with a modest amount of data. For the selection of this signal we will thus rely on lepton triggers. One of the main questions that needs to be addressed is how efficient the various trigger items are. Since generally the trigger efficiency directly influences any cross-section measurement, we need to measure it with the smallest possible uncertainty. Data-driven methods, which are Monte Carlo independent and have the advantage of automatically taking into account unknown aspects that are difficult to simulate, are under extensive study. The Tag&Probe method is used to evaluate lepton-trigger efficiencies from real data using double-object final states, such as Z→µµ. Furthermore the trigger efficiency varies as a function of energy, position and isolation of the triggered object. Since Z→µµ and tt events have very different signatures, the lepton trigger efficiency should be a parameterized function of kinematical as well as topological variables. The Tag&Probe lepton-trigger efficiencies for the ATLAS detector and their extrapolation to tt events are presented.

100% | Mobile Layout | Deutsche Version | Contact/Imprint/Privacy
DPG-Physik > DPG-Verhandlungen > 2010 > Bonn