Search for jet-jet resonances in association with a leptonic W decay at the ATLAS experiment

The analysis presented in this thesis addresses the study of two processes that share the same signature: the diboson $WW/WZ$ semileptonic decay and a jet-jet resonance with an invariant mass equal to $145~GeV/c^{2}$ produced in association with a $W$. In both cases the leptonic decay of the W is identified as an energetic electron or muon along with missing transverse energy forming a transverse mass~($M_{T}$) consistent with that one produced by the $W$ boson. The hadronically decaying $W/Z$ or the resonance at $145~GeV/c^{2}$ are searched in the jet-jet invariant mass distribution after a selection based on the event kinematics. The resonance at $145~GeV/c^{2}$ has been observed in April 2011 by the CDF detector at the Tevatron collider with a significance of $4.1$ standard deviations and is not predicted by Standard Model. Therefore it is important to determine the presence of this signal in other experiments and this thesis investigates the possibility to reveal the same resonance in data collected by the ATLAS experiment in the first half of 2011. Even if production conditions at LHC are different than those ones at Tevatron, it has been chosen to use the same selection applied by CDF in order to avoid any assumption on the type of signal process. In my thesis I have analysed about $1~fb^{-1}$ of ATLAS data and reproduced the CDF selection to obtain the jet-jet invariant mass distribution. This distribution has been compared to that one obtained from the Standard Model prediction. This information has been used to identify the region where data exceed Standard Model expectation with the highest significance. The study reports no evidence of excess in the jet-jet invariant mass. The $WW/WZ$ cross-section has been measured in ATLAS in the leptonic channel and not yet in the semileptonic one. These studies have twofold purpose: the definition of a sample that would allow to set constraints on triple gauge couplings~(TGCs) and test the goodness of Monte Carlo reproduction of the $WW/WZ$ signal, an important background in Higgs searches. The triple gauge couplings are the constants which determine the strength of the interaction among three vector bosons and enter in the $WW/WZ$ cross-section. The Standard Model predicts the TGCs value which may be tested with the measurement of $WW/WZ$ cross-section. The thesis focuses on the development of two selections that reduce and model the background yields in order to isolate the $WW/WZ$ signal allowing its measurement. Specific cuts on jet kinematic reduces the main backgrounds constituted by associated production of $W$ and jets, $t ar{t}$ and multijet QCD. One study aimed at reaching high statistical significance~($sim 3$) in the $1.33~fb^{-1}$ of data analysed, therefore the cuts are chosen evaluating which ones lead to the best $S/sqrt{B}$ ratio. The other study focuses on the improvement of the $S/B$ ratio and consequently tries to raise the signal significance with respect to the systematic error. Finally, the effects of the dominant systematics are discussed and control regions are studied to verify the goodness with which the Monte Carlo simulations reproduces the main backgrounds.