Development of an algorithm to increase the KS tracking efficiency at Belle II

Sumitted to PubDB: 2025-04-28

Category: Master Thesis

Tags:

Principal Authors Matilde Carminati
Date 2024-07-22
Belle II Number BELLE2-MTHESIS-2025-006
Abstract The objective of my thesis is the development of an algorithm to improve the tracking efficiency of the KS at the Belle II experiment. In many decay of interest for the Belle II physics program, in particular for high-precision measurements of CP violation, the KS is often among the produced daughter particles hence the importance of its reconstruction. The Belle II tracking system provides excellent efficiency in reconstructing charged particles produced close to the e+e- interaction region. However, particles generated far from it are more challenging to reconstruct; the reconstruction efficiency indeed decreases as the distance of the mother's decay vertex from the interaction point increases. The charged pions produced in the KS decay fall into this particles’ category. For example, in BBbar events, the reconstruction efficiency of KS decaying near the interaction point is around 85%, but it drops below 60% when the decay vertex is more than 10 cm away from the interaction point. The primary source of inefficiency is the failure to reconstruct one of the two daughters. By recovering all these cases, a maximum increase of 15% in the total efficiency of the KS could be achieved, resulting in an equivalent increase of luminosity of 20%, for the measurements in channels with at least one KS in the final state. The main topic of this thesis is the development of a new algorithm to recover KS for which only one daughter is correctly reconstructed and the other one is missed by the standard tracking algorithms. In particular, the goal is to find the track of the KS missing daughter, utilising information only from the Silicon Vertex Detector (SVD) and the fitted track of the reconstructed daughter. The position of the KS vertex can be estimated to lay on the helix arc of the reconstructed pion between its innermost SVD hit and its intercept on the previous SVD layer. This estimate is used together with the assumption that the KS is coming from the interaction region to solve a second-order equation from the quadri-momentum conservation, and to compute two estimates of the KS momentum, and thus the momentum of the missing pion. Each estimate is used as seed, together with the estimated KS decay vertex position, to build a temporary helix of the missing pion. Such helix is used to search for SVD hits near its intersections with the SVD layers, in order to reconstruct the track. The algorithm is implemented from scratch in the official Belle II software framework to explore its potential. The first results on the overall efficiency indicate that more work is required to optimize the algorithm and obtain a larger impact on physics. The main source of inefficiency has been identified and possible improvements are discussed.

Files