Another Approach to Track Reconstruction: Cluster Analysis

Abstract

A novel combination of data analysis techniques is introduced for the reconstruction of primary charged particles and of daughters of photon conversions, created in high energy collisions. Instead of performing a classical trajectory building or an image transformation, efficient use of both local and global information is undertaken while keeping competing choices open. The measured hits in silicon-based tracking detectors are clustered with the help of a k-medians clustering. It proceeds by alternating between the hit-to-track assignment and the track-fit update steps, until convergence. The clustering is complemented with the possibility of adding new track hypotheses or removing unnecessary ones. A simplified model of a silicon tracker is employed to test the performance of the proposed method, showing good efficiency and purity characteristics.