Track Simulation and Reconstruction in the ATLAS experiment

The reconstruction and simulation of particle trajectories is an inevitable part of the analysis strate- gies for data taken with the ATLAS detector. Many aspects and necessary parts of a high-quality track reconstruction will be presented and discussed in this work. At first, the technical realisation of the data model and the reconstruction geometry will be given; the reconstruction geometry is charac- terised by a newly developed navigation model and an automated procedure for the synchronisation of the detailed simulation geometry description with the simplified reconstruction geometry model, which allows a precise description of the tracker material in track reconstruction. Both components help the coherent and fast integration of material effects in a newly established track extrapolation package, that is discussed in the following. The extrapolation engine enables a highly precise trans- port of the track parameterisation and the associated covariances through the complex magnetic field and the detector material. As a direct consequence, the track parameter resolution could be improved — in particular at the low momentum regime where material effects are most pronounced. This will be shown in the context of the new track reconstruction chain that is presented together with an extensive validation section. The intrinsic navigation model and modular design of the extrapolation engine are also key aspects of a new fast Monte Carlo track simulation engine that is described in the second part of this thesis. It includes the most relevant physics processes, such as the integration of random multiple scattering and energy loss effects when the particle traverses the detector; also particle decay and the simulation of photon conversions are included in the new fast simulation. It reaches high compatibility with the full simulation, while decreasing the execution time per event by about 2 orders of magnitude