The CMS muon high-level trigger at HL-LHC: performance and optimisation

The High-Luminosity upgrade for the Large Hadron Collider at CERN will increase its instantaneous luminosity up to a factor of 7.5 over design. This poses significant challenges for the experiments both from the hardware and software points of view. The detectors will be subject to much more radiation, requiring careful consideration of the radiation hardness of the to-be-installed detectors and the damage accumulated by the current ones. From the software point of view, data acquisition and physics reconstruction will have to cope with hundreds of superimposed proton-proton collisions at every bunch crossing. Therefore, all the experiments will undergo a major upgrade to improve their detector, triggering solutions, and reconstruction software. This Master's Thesis presents the foundations for a novel approach to Muon reconstruction at the CMS experiment. Since the muon tracking algorithms have demonstrated excellent performance during previous Runs, the new algorithms aim to maintain or improve the current physics performance, while reducing the computational load. This is achieved by taking full advantage of the upgraded first-level hardware trigger and eliminating some of the redundancy present in the current reconstruction workflow. Both physics and computing performance have been evaluated using Monte Carlo simulated samples targeting expected HL-LHC conditions. The results show close-to-current physics performance with up to about 40% improvement in the timing of some of the modified reconstruction modules. Furthermore, the number of fake tracks produced is also reduced by about 30-40% depending on the reconstruction step, decreasing the complexity of the reconstruction as a whole.