Upgrade of the global muon trigger at the CMS experiment

The Compact Muon Solenoid (CMS) experiment is one of two general purpose detectors at the Large Hadron Collider (LHC) at the particle physics research laboratory in Geneva (CERN). As such it allows a broad array of physics analyses from precision measurements of the standard model of particle physics to searches for exotic new particles. A series of upgrades and maintenance procedures took place in the first shut down from 2013 to 2015. The aim was to prepare the LHC for the collision energy of 13 TeV and further increase its luminosity. During this shut down also upgrades of the CMS experiment were installed.Due to the high rate of collisions at the LHC, it is impossible to record all such events. In order to reduce the event rate to a manageable level, a trigger system is deployed that selects interesting events. At the CMS experiment this system is divided into two levels: A first hardware based system that is optimised for speed and a second that is software based and applies more time consuming and precise algorithms. Due to the changed, more challenging conditions after LHC upgrade an upgrade of the first level of the trigger system is essential.All three muon detectors of CMS send information to the trigger system. This work describes the design of the Micro-Global Muon Trigger (uGMT), a central part of the muon trigger that receives all muon candidates of a collision event. The main part of the algorithm is the selection of the best candidates that are sent to the final component of the trigger. Since the components that reconstruct these candidates do not exchange information, candidates that are found twice need to be found and eliminated. Furthermore, the uGMT receives measurements of energy deposits from the calorimeters. With this information an isolation variable can be introduced that allows to suppress events from background processes.This thesis presents results showing that such an isolation variable can be used to suppress the rate of events by a factor of two while maintaining the same acceptance for signal processes. A selection based on quality criteria of the muon candidates can be used to further decrease the rate by a factor of three which leads to loss of a few percent of efficiency for signal events.The results were derived from a software-program that simulates the behaviour of the electronic system. The implementation of the program that simulates the uGMT algorithm was developed as part of this thesis. The simulation was also used to verify the behaviour of the hardware realisation.The prerequisite for successful data taking is a reliable monitoring and configuration system that detects problems early on and allows react correspondingly. The thesis presents contributions to the software development of this system for the first level trigger.