1 / 19

EMC session, Elba general Meeting, May 31, 2010

Had Breco analysis: E extra studies. Elisa Manoni INFN Sez. Perugia. EMC session, Elba general Meeting, May 31, 2010. Outline. Aim of the study Analysis strategy and validation Background gamma characterization E extra and different background scenario Conclusions.

swansonm
Télécharger la présentation

EMC session, Elba general Meeting, May 31, 2010

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Had Breco analysis: Eextra studies Elisa Manoni INFN Sez. Perugia EMC session, Elba general Meeting, May 31, 2010

  2. Outline • Aim of the study • Analysis strategy and validation • Background gamma characterization • Eextra and different background scenario • Conclusions

  3. PacHadRecoilUser and Breco reconstruction • SemiExclusive reconstruction of Hadronic B modes • limit the number of reconstructed Breco channels according to their purity (>50%). Breco mode classification: neat : purity > 80% clean : 50% < purity < 80% dirty : 8%<purity<50% • Available Bsig modes • K* • K , Ks() •  , with e, , , (0), a1()

  4. Eextra shape and background mixing February production analysis: • mix physical events with Bhabha + Radiative BhaBha (integrate barrel EMC signal in 1ms time window) • background (radiative BhaBha) dramatically increase the number of reconstructed neutrals  in signal MC Eextra shifts at high values, loosing the bin 0-discriminating power • 1ms time window found to be too conservative, 0.25ms more realistic • how do the Eextra shape change by reducing the integration time (and the bkg) to 1/4? Fast Sim B+K+ signal MC Full Sim B+K+(Ksp) signal MC generics

  5. Analysis strategy • Use February production tuples with background mixing switched on: • B+K+ signal MC • B+B- and B0B0bar generic MC • compute Eextra at ntuple level (gamma block) • remove gammas overlapping with Breco or Bsig • sum extra-gamma energy if Eg>Egmin • ¼ background scaling • identify gammas from background by using mctruth infos • associate to each extra-gamma (in the barrel emc) a random number ∈ [0,1] • if randnum>0.25, reject the extra-gamma , otherwise use it to compute Eextra • will consider 3 bkg scenario • 100% bkg  no bkg reduction • 25 % bkg  ¼ bkg reduction • w/o bkg  all bkg gammas removed

  6. Analysis validation: Offline Eextra computation (I) • preliminary study (some EMC meeting ago): not perfect matching between Eextra’s computed at reconstruction level and offline histo Eextra_ntuple Eextra_offline

  7. Analysis validation: Offline Eextra computation (II) • some minor code issues • emc coverage (cosqlab ) : • FastSim and FullSim bwd [-0.9782, -0.8965], brr+fwd [-0.777,0.9625] • PacHadRecoilUser bwd [-0.9655,-0.8581], brr+fwd [-0.7859,0.9622]  apply same cosqlab cuts as in PacHadRecoilUser • duplicate candidate bug: at ntuple level the ithU cand owns the UsrVariables (mES, DE, Eextra…) of the first  choose the first U as best candidate (bug now fixed in V0.2.3 + updated Patches_devel packages)

  8. Analysis validation: Offline Eextra computation (II) • some minor code issues • emc coverage (cosqlab ) : • FastSim and FullSim bwd [-0.9782, -0.8965], brr+fwd [-0.777,0.9625] • PacHadRecoilUser bwd [-0.9655,-0.8581], brr+fwd [-0.7859,0.9622]  apply same cosqlab cuts as in PacHadRecoilUser • duplicate candidate bug: at ntuple level the ithU cand owns the UsrVariables (mES, DE, Eextra…) of the first  choose the first U as best candidate (bug now fixed in V0.2.3 + updated Patches_devel packages) histo Eextra_ntuple Eextra_offline

  9. Analysis validation: scaling algorithm • B+B-_generic sample, without bkg mixing 100% bkg -75% bkg no bkg

  10. Background gamma multiplicity and angular distribution • 100% background B+B- generic MC B+K+ signal MC gmultiplicity bwd region barrel region fwd region g cosqlab _ bkg g _ phys g

  11. Background gamma energy spectrum • 100% background B+B- generic MC B+K+ signal MC Eg (GeV) Eg (GeV) EgTOT (GeV) EgTOT (GeV)

  12. Eextra barrel vs Eextra forward • 100% background B+B- generic MC B+K+ signal MC FWD EMC BARREL EMC

  13. Eextra as a function of Egmin • 100% background B+K+ signal MC Egmin = 70 MeV Egmin = 50 MeV Egmin = 30 MeV B0B0bar generic MC B+B- generic MC

  14. Comparing bkg scenarios: Eextra Egmin = 30 MeV w/o bkg 25% bkg 100% bkg B+K+ signal MC B0B0bar generic MC B+B- generic MC

  15. Comparing bkg scenarios: gamma cosqlab w/o bkg 25% bkg 100% bkg bwd region barrel region fwd region B+K+ signal MC B+B- generic MC B0B0bar generic MC

  16. Eextra after B+K+ selection • cut and count selection a-la-BaBar, removing requirements on extra p0 and Eextra B+K+ signal MC Egmin = 30 MeV w/o bkg 25% bkg 100% bkg

  17. Conclusions • Eextra computation and scaling algorithms seem to work properly • Egmin changes Eextra shape, cut at 70 MeV make signal MC shape peaky at low Eextra even with 100% background • scaling the number of not matched extra-gamma to 25%, Eextra has peaky shape in signal MC; both in signal and generics Eextra range is enlarged: need to properly define a new signal region wrt BaBar • Further investigation by using next production samples • generate ad-hoc BB cocktail to provide high statistics sample where Eextra shape for BB generic after full selection can be studied (see tomorrow DGWG session)

  18. Back-up slides

  19. B+K+: selection efficiency • BaBar-like cut and count analysis Qtag = ± 1 5.270<mES<5.288 GeV/c2 |cosBreco,Thrust|<0.85 K candidate from Bsig |cos*trk|<0.85 NextraTrk=0 Eextra<0.4 GeV N0=0 pKB > 1.1Gev/c -0.85<cospmiss<0.9

More Related