Thesis BELLE2-PTHESIS-2017-001

Development of Pattern Recognition Algorithms for the Central Drift Chamber of the Belle II Detector

Viktor Trusov ; Michael Feindt ; Andreas Meyer

2016
Karlsruhe Institute of Technology Karlsruhe

Abstract: In 1973 Kobayashi and Maskawa proposed a mechanism which describes CP violation by introducing an irreducible complex phase in the quark mixing matrix. They predicted three families of quarks that was confirmed by the later discovery of the b quark. The problem of CP violation can shed light on the problem of matter-antimatter asymmetry, which is one of the mysteries of modern cosmology. Also, B physics and CP violation studies have a potential to measure parameters of the electroweak interaction and to test the Standard Model. The studies of CP violation may lead to physics beyond the Standard Model and to New Physics Phenomena. This brings B physics studies to the level of the most important objectives in modern high energy physics. Experiments which are focused on the studies of B physics are called B factories. They are designed to generate a unprecedentedly high amount of B mesons and to study their properties. The process e+e− -> Υ(4S) -> BB is responsible for generation of B mesons at B factories, and it brings clean physics environment in which only final state particles from B decays are registered by the detector. The Belle II experiment will be one of B factories with the highest luminosity ever achieved. It was developed as an upgrade of its predecessor, the Belle experiment, but with redesigned components and currently being under construction. The experiment is intended to analyze generated B mesons by combining measured parameters of the decay products, such as the type of particles in the final state and their kinematic parameters. The Belle II detector itself can only provide response signal, generated by the decay products. The task of event reconstruction is entrusted to the reconstruction software, which processes the signals from the detector. Pattern recognition (which is also called tracking in scope of high energy physics experiments) plays a substantial role in the event reconstruction. It is intended to provide a sufficient amount of information about charged particles, created in studied processes, for further physics analyses. As it is a crucial part of the reconstruction, it should operate with high efficiency. Inefficiency in tracking leads to losses in the event information, and as a result physics processes would be wrongly reconstructed or discarded. In scope of the Belle II experiment, pattern recognition should correctly reconstruct as much as possible events to preserve collected data, and withing limited computational time since a large amount of data should be processed. An overview of the Belle II experiment is given in chapter 2. Still, the detector is not deployed, and the first physics run is expected to take place in 2018. Thus, the Belle II software framework is developed and tested using Monte Carlo simulated data. The description of the software framework is presented in chapter 3. Most of its features were used to develop the pattern recognition algorithm, which is discussed in present work. In chapter 4 a general overview of tracking techniques is given. Chapter 5 contains discussion of one of the tracking detectors of the Belle II , the Central Drift Chamber, and algorithms which have been chosen to perform pattern recognition using its output. The main matter of the current work is given in chapters 6 and 7, where the essential features of the developed tracking algorithm and its performance are presented. The algorithm is based on the track segment recognition technique introduced by the ATLAS experiment, where Legendre transformation is used to find patterns of hits which are sharing the common tangent. The feasibility of the method for purposes of the Belle II tracking has been tested in work and proved that Legendre based tracking is capable to efficiently reconstruct tracks with low execution time. The current work is a continuation of the former development, aimed to improve the capability of the algorithm to reconstruct generic BB decays. Variuos features have been introduced to cover a wide range of the kimematic parameters of tracks of different types. The performance of the method was studied in comparison to the Belle tracking. Also, the stability of the pattern recognition was tested in conditions of high beam induced backround levels.

Note: Presented on 04 11 2016
Note: PhD

The record appears in these collections:
Books, Theses & Reports > Theses > PhD Theses

 Record created 2017-04-24, last modified 2017-04-24


Fulltext:
Download fulltext
PDF

Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)