When it comes to particle physics experiments, we need to be able to reconstruct processes that happen in our detectors. For that we use detectors capable of identifying charged particles, one of such detectors at Belle II is aerogel RICH, which is mostly used for discriminating between kaons and pions in the whole kinematic region of the experiment.
To understand what is happening in our detectors and the measurements that we get, we use simulations which recreate measured data and require calibration to get a good agreement between simulated and measured data. We are able to improve certain aspects of simulations by changing parameters used in them, which is part of this thesis' goal. Second thing we focused on, is improving background contribution for calculating PID likelihoods used by ARICH reconstruction algorithm, in order to achieve better agreement with observed data. We got good results in both of the tasks we set out to do, using common calibration data from $e^+e^- \rightarrow \mu^+\mu^-$ events as well as from reconstructed proton tracks to observe background contributions. For protons, we used tracks below Cherenkov radiation threshold and with them being uncommon for detector calibration studies, we had to reconstruct them ourselves. For each part of the calibration we observed improvement for both different particle types. Lastly, we checked particle identification performance for discriminating between kaons and pions and observed improvement after calibration.
|
