Abstract
| Online track reconstruction is an important ingredient for event selection at Large Hadron Collider (LHC) experiments. In the ATLAS experiment the first stage where this goal will be achievable is the software-based Second Level Trigger (LVL2). In this contribution we present an algorithm for fast pattern recognition and reconstruction of charged tracks and of the primary vertex in the framework of the High Level Trigger (HLT) of ATLAS. The pattern recognition makes extensive use of Monte Carlo Look Up Tables to quickly identify, in the innermost layers of the ATLAS silicon detectors, triplets of space points reconstructed from hits produced by the same track. The reconstruction strategy is compared, in the ATLAS LVL2 framework, with an alternative tracking algorithm, showing the complementarity of the two approaches. The algorithm’s performance is presented for different event topologies and luminosities, showing good tracking capabilities and uniform results with mean execution times which are compatible with the LVL2 requirements. |