This web page provides distributions and tables of a detailed study performed to develop a B flavor tagging algorithm, based on the charge correlation between the signal B meson and the fragmentation tracks produced nearby. Based on Monte Carlo events, we evaluate the performance of different Same Side Tagger (SST) algorithms on B+, Bd and Bs decay modes. We develop and optimize a tagging algorithm using kaon identification and dilution parameterization.
This is the web page summarizing results of the Same Side (Kaon) Tagger analysis. These results were produced as part of the CDF ongoing effort in a Bs mixing analysis.
A complete version of the results can be found in the CDF note 8206
Below are the eps versions of tables from the note meant for downloads.Final results
Systematic uncertainties
Below are the eps versions of figures meant for downloads.
Data-MC comparison of average dilution
Data-MC comparison of parameterized dilution
Resonances / V0's
Data-MC tagging tracks multiplicity before and after PID > 1 cut
Data-MC tagging tracks pt before and after PID > 1 cut
Data-MC tagging tracks ptrel before and after PID > 1 cut
Data-MC tagging tracks plrel before and after PID > 1 cut
PID information
Predicted dilution as a function of the tagging track transverse momentum
Predicted dilution as a function of PID
Variations of some Monte Carlo distributions with respect to data
Lund variations for systematic studies
Systematic uncertainties in Monte Carlo compared with data values
Additional Material:
HEPG ancestors
The following plots show the anchestor particles of kaons, protons and pions in a cone of 0.7 around B meson candidates. This study has been done on generator level, no reconstruction cuts are applied on the tagging track candidates but a minimum transverse momentum of 0.45 GeV/c is required. The white and red contributions indicate the charge correlation to the B meson. Lambda's have a very long lifetime thus there decay products are in 95% of the cases not anymore part of the generator level information but are generated by the detector simulation, therefore lambda's don't show up as proton ancestors in those plots. Lambda anchestors are shown seperately, about half of them decay into proton-pion.HEPG dilution dependence on the transverse momentum of the tagging track
In the following we select tagging tracks using the maximum longitudinal momentum algortihm. Again no reconstruction criteria but the minimum pt cut have been applied. The dilution is sperately plotted for the case where the selected tagging track is a kaon, proton or pion. Those plots demonstate that in the case of B_s mesons all the tagging power come from kaons.Efficiency and dilution dependence on B** fraction for max plrel algorithm
The following plots show the efficiency and dilution as a function of B** content in the Monte Carlo (black lines). The overlaid red line indicate the +/- 1 sigma range of the according numbers in data. The data is well compatible with a fraction of 20% of B+, B0 coming from excited B decays. A variation of [16%,35%] for the systematic studies has been chosen.