Several interesting physical processes lead to the production of low-energy (soft) b-quarks in the final state. The produced b-hadrons are also low energy and not bound to reconstructed jets. To select such processes, low-energy b-hadrons must be identified independently of the reconstructed jets. Standard FTAG algorithms are limited to the phase space inside reconstructed jets (typical cone size ~0.4 and less) and therefore are not able to detect soft B-hadron decays with a much wider decay product distribution. A tool for the inclusive reconstruction of all secondary vertices in an event has already been developed. The reconstructed B-hadron decay secondary vertices can be either associated with jets to improve the jet b-tagging efficiency itself (a functionality already exploited by CMS) or used independently of jets for an event with B-hadrons tagging/rejection.
The principle QT deliverables for Beatrice Cervato should be the optimisation of the selection efficiency and background suppression of the soft B-hadron vertex reconstruction tool and validation of its performance on data. The fake-vertex rate of the inclusive tool will be compared to the fake rate of the standard b-tagging vertex reconstruction inside jets. This should improve the understanding of the b-tagging fake rate and facilitate the light-jet rejection calibration.
Regular reports on the status will be given in the FTAG algorithm meeting. The work should be documented in an internal note and the software should be maintained in a git repository.