Track Reconstruction in the Forward Region of the Detector ILD at the Electron-Positron Linear Collider ILC

The subject of this thesis is the reconstruction of charged particle tracks in the forward region of the International Large Detector (ILD), one of two validated detector concepts for the future International Linear Collider (ILC). Recent results from the Large Hadron Collider (LHC) suggest that the last missing piece of the Standard Model (SM) of particle physics, the Higgs boson, could have been found. Complementary to the LHC, an electron-positron linear collider would have the capability to explore with high precision the characteristics of the Higgs boson (e.g. spin, parity, coupling strengths), to compare them against the predictions of the SM, and also give unbiased contributions to the search for physics beyond the SM, such as supersymmetry or extra spatial dimensions. Experiments at ILD will benefit from this detector's tracking system, consisting of a large central time projection chamber (TPC) augmented by several silicon tracking systems, granting unprecedented track resolution, redundancy and angular hermeticity. The forward region of ILD, covering the space between beam tube and TPC, contains the Forward Tracking Detector (FTD): two arms of seven disk-shaped silicon detectors (two Si pixel and ve double-sided Si strip detectors). In order to fully exploit ILD's hardware assets, the software for event reconstruction must aim for both the highest level of precision and efficiency. In this context, new software for track reconstruction in the forward region was developed by the author (packages KiTrack and ForwardTracking): it processes the signals of FTD with the goal to efficiently nd and precisely reconstruct the genuine tracks that have traversed FTD and caused these signals. The methods used are based on state-of-the-art algorithms: a cellular automaton (CA), a Kalman lter (KF), and a Hopeld neural network (HNN). The new packages follow a modern object-oriented design philosophy, granting high flexibility and maintainability. The results show superior performance w.r.t. older legacy software, yielding higher efficiencies and better handling of the expected background concerning ghost rate, efficiency and processing time. The forward tracking packages presented in this thesis have been successfully implemented into the standard event reconstruction framework of ILD. They are currently used for benchmark event processing for the Detailed Baseline Design (DBD), a report outlining the feasibility and features of the International Large Detector, to be published around the end of this year.