Abstract : The subject of this thesis is the evaluation of the discovery potential of the ATLAS detector at the Large Hadron Collider for the Standard Model Higgs boson in vector boson fusion (VBF) production and a subsequent decay into a τ -lepton pair (H → τ ^+ τ^ −). This is one of the most promising discovery channels in the low mass range, which is the mass range favored from precision measurements of the electroweak interaction. The decay modes where both τ leptons decay leptonically and where one τ lepton decays leptonically and the other one hadronically were studied. The characteristic vector boson fusion topology, which consists of two jets in the forward regions of the detector and the Higgs boson decay products in the central region, provides a unique signature allowing the suppression of background. In addition, since vector boson fusion is a purely electroweak process, no QCD activity is expected and thus no central jets are expected for signal events. This allows the central jet veto cut application to further reject background processes. A cut-oriented analysis was used, focusing on the central jet veto cut optimization. The main ob jective was to investigate the Higgs boson ATLAS discovery potential with an integrated luminosity of 30 fb^−1 in the mass range 115 GeV ≤m_H ≤140 GeV, for an LHC energy at the center of mass √s=14 TeV. After the application of all cuts, an excess above 5σ of signal signiﬁcance was found for the mass range 115 GeV ≤m_H ≤125 GeV. In H → τ^ + τ^ − decay channel, transverse missing energy (E_T ) resolution is of high importance since it affects directly the resolution of the Higgs boson mass. This was the initial motivation for performing and presenting in this thesis a calorimetric commissioning study based on E_T quantities, focused on the electronics noise E_T contribution, measured with cosmics data of the period fall 2008. A Gaussian behavior of noise in all liquid argon calorimeters was found, whereas a region of non-Gaussian tails in scintillating tile calorimeter was high-lighted. For the latter, a new noise model was tested using a double Gaussian parameterization resulting in a more realistic description. Finally, this analysis provided a useful calorimeter commissioning tool, which allowed the observation and the correction of several features in the ATLAS calorimeter behavior.