Performance of $\tau$ lepton identification in ATLAS 7 TeV data

The reconstruction and identification of $ au$ leptons are important in physics analysis of Standard Model (SM) processes and searches for new phenomena. In this paper the comparison of distributions of identification variables between data and Monte Carlo samples is done, using the initial dataset collected by the ATLAS detector at a centre-of-mass $sqrt{s} = 7$~TeV, corresponding to an integrated luminosity of 244 nb$^{-1}$. As the number of recorded $ au$ leptons is small, the background jets reconstructed as tau-leptons candidates can be used to check the algorithms performance. The background efficiency suppresion is tested on QCD dijet data sample using cut-based $ au$ identification criteria. The multivariate methods, such as boosted decision trees and projective likelihood, are also tested to check their background rejection performance.