Low-momentum tracking in the Belle II tracker

The Belle II experiment at the future SuperKEKB collider is an upgrade of the Belle experiment at KEKB in Japan. The SuperKEKB collider is the successor to the asymmetric electron-positron KEKB collider, aiming for a 40-fold increase in luminosity. One of the strengths of the upgraded experiment is that it can probe physics beyond the standard model. The main tool in this search for new physics are flavor changing neutral current processes. A natural place to investigate a wide range of such processes is the decay of B-mesons. In order to take full advantage of the luminosity delivered to the experiment, the detector and the reconstruction software must be optimized for reconstructing the various B meson daughters. The reconstruction of low-momentum pions is a crucial ingredient in this program. It is the aim of this project to improve track finding and track reconstruction in the silicon vertex detector (SVD) of Belle II, down to the lowest transverse momentum possible of about 50-60 MeV/c, in order to significantly increase the effective data sample and the physics potential of the experiment. In the course of the project the necessary algorithms for finding and estimating low-momentum tracks will be developed, without using external information from other tracking detectors. An important aspect of the reconstruction algorithms will be the correct treatment of material effects such as multiple Coulomb scattering and energy loss by ionization at very low particle energies. Finally, the algorithms will be designed to be robust, i.e., as insensitive to background as possible. In order to reduce the data rate from the Belle II pixel detector , the SVD will also be used for on-line data reduction. The idea is to reconstruct tracks in the SVD, and to extrapolate them into the pixel detector. Around each intersection point a region of interest is formed, and only pixels in such a region of interest are read out. As a consequence, track reconstruction in the SVD must be capable of being performed in real time, either on specialized hardware or on a farm of commodity processors. Therefore, the algorithms developed and implemented in the course of the project have to be fast enough to fulfill the timing requirements of the pixel readout. The global track finding strategy has been formulated and several promising candidate algorithms have been identified. The latter will be further developed and adapted to the specific needs of low-momentum tracking. They will be optimized and validated using simulated data, and a comparative evaluation will be carried out. If the schedule of the accelerator and the detector upgrade permits, the project will be concluded by tests with real data.

The experiment Belle II is currently being installed in the KEK laboratory in Tsukuba, Japan. The main physics goal of the experiment is the study of CP violation, a fundamental asymmetry in the interactions of matter and antimatter. A new accelerator, an electron-positron collider named SuperKEKB, is now in its commissioning phase at KEK. In its two rings, it stores, accelerates and finally brings to collision electrons and their antiparticles, positrons. The energy of the two beams is adjusted such that a pair of B mesons is generated with high probability in each collision. The asymmetry mentioned above can then be observed and studied in the decays of the B mesons. As B mesons travel only a fraction of a millimeter before they decay, they cannot be observed directly. Instead they have to be reconstructed from their decay products. In a very important decay mode, one of the decay products is a pi meson (pion) that can be measured in the silicon vertex detector of the experiment. As these pions have rather low momentum, special pattern recognition algorithms are required in order to find the signals generated by the pion, so that its properties such as charge, direction and momentum can be estimated. In the course of the project methods were developed that solve the problem with sufficient efficiency and precision. As the experiment is still in the construction phase and cannot yet deliver real data, the algorithms were evaluated mainly on simulated data. In addition, they were deployed in a beam test experiment and successfully tested on real data. They are implemented in the software framework of the Belle II experiment and are an integral and important part of the software that analyses the data recorded by the experiment. They are thus an important contribution to the success of the experimental program of Belle II, which will start taking data in 2018.