Di-muon trigger efficiency

Document Sample
Di-muon trigger efficiency Powered By Docstoc
					 LVL1, LVL2 low pT muon triggers efficiency calibrations
with J/Y Upsilon and non resonant di-muons from B-events

                  Hironori Kiyamura (Kobe)
           Takanori Kono, Cristobal Padilla (CERN)
                  Daniel Scheirich (Prague)
        Shlomit Tarem, Natalia Panikashvili (Technion)

   LVL1 di-muon trigger efficiency calculation
   • Introduction
   • Appendix
   • Single muon trigger efficiency
   • Di-muon trigger efficiency
   • Summary and outlook

   Status of CSC note 07
   • Contents of the note
   • Status of the note

2007/09/03                  B-physics meeting    2
LVL1 di-muon trigger efficiency calculation

   • Estimate Di-muon Trigger Efficiency using real data.
             – Identify pre-scaled trigger is needed for this study.
   • Measure single muon trigger efficiency using J/Y -> mm process.
   • From the result, make Trigger Efficiency Map for each (pt,h,f).
              Single muon trigger efficiency will be obtained from this map.
   • Calculate di-muon trigger efficiency as the products of two muon’s
     trigger efficiency.
   Data samples (AOD,ESD)
   • Signal
             – Trig1_misal1_mc12.017516.PythiaB_bb_Jpsimu6mu4X.v12000602
             – Trig1_misal1_mc12.017503.Pythia_directJpsimu6mu4.v12000602
   • Background
             – trig1_misal1_mc12.017500.PythiaB_bbmu6X.v12000602
             – trig1_msial1_csc11.0050(09~16).J(0~7)_pythia_jetjet.v12000601
2007/09/03                                   B-physics meeting                  4
Appendix 1

 LVL1 trigger configuration                                         900GeV configuration
                                                                    Normal configuration

       ENDCAP                                             BARREL

  900GeV trigger configuration is slightly different with normal one.
   1MU06 trigger efficiency is different between these two configurations.

2007/09/03                            B-physics meeting                                5
Appendix 2

 Muon reconstruction efficiency

                                                                 The difference is large
                                                                 for low pt muons.

                                                    ▲   StacoCombined muons (triggered)
                                                    ▲   StacoCombined muons (not triggered)
                                                       MuonBoy muons (triggered)
                                                       MuonBoy muons (not triggered)

 Reconstruction efficiency depends on whether muon is triggered or not.
               We need to understand the correlation between trigger and offline.
               We will measure LVL1 trigger efficiency for reconstructed muons.
2007/09/03                          B-physics meeting                                      6
Single muon trigger efficiency measurement

 Leg technique                      Step.1       seed muon(MU06)

                                                                         dR  dh 2  df 2
                                      probe muon                      Trigger judgment

                                J/Y                                   dR matching between
                        (2.88GeV<Mmm<3.3GeV)                          ROI and Hits on track.
                                                                      -> dR<0.1 -> triggered

 •     Use only MU06 triggered events and Muonboy reconstructed muons
        1. Look for MU06 triggered muon in both barrel and endcap
            -> seed muon
        2. Look for another muon of decay products from J/Y -> probe muon
        3. Using the probe muon, single muon trigger efficiency is calculated.
 •     With this method, single muon trigger efficiency is calculated and
       validated by comparing with single muon samples.
2007/09/03                               B-physics meeting                                     7
Single muon trigger efficiency for pt                          Obtained by J/Y -> mm samples
                                                               Single m (reconstructed by Mboy)
                                                               Single m (all muons)

      ENDCAP                                                  BARREL

                 Fit by
                      1  exp (  a(pt  b))

• No bias exist in the results of single muon samples.
    For reconstructed muons, the efficiency from data (black square) matches
      well with that of single muon sample (red square).
    This method works well !!
※ Discrepancy is observed in single muon efficiency by requiring offline
  reconstructed muons. (mentioned in slide #5)
                          Next -> Divide this result to make trigger efficiency map.
2007/09/03                                B-physics meeting                                   8
Single muon trigger efficiency for each (h,f) region (BARREL)

 Barrel is divided into 10 regions for h and f respectively.
F=-3.14                                                                F=3.14

 In some bins, points below 4GeV are missing and the fit doesn't work quite well.
 -> Need sample with mu6+mu0
2007/09/03                             B-physics meeting                            9
Single muon trigger efficiency for each (h,f) region (ENDCAP)

 8 octants are merged and divided into 10/6 regions for h/f respectively.

    IDf = 0                                                          IDf = 5

2007/09/03                            B-physics meeting                           10
   Trigger efficiency map

         Fitting parameters are recorded with the center position of each region.
          Recorded values are linearly extrapolated to all over Endcap/Barrel.
          A                            a                           b

          A                            a                           b

                                 (A,a,b) values at (h,f) are obtained from these maps.
   2007/09/03                               B-physics meeting                       11
Di-muon trigger efficiency calculation

   • Single muon trigger efficiency : m1(pt1,h1,f1), m2(pt2,h2,f2)
     (These two values are obtained from trigger efficiency map)
   • Di-muon trigger efficiency : mm = m1×m2
    Trigger efficiency for J/Y events can be calculated as following,
      – Define q as the decay angle of muons from J/Y and P as the
         momentum of J/Y.

             CM             m1
                                              LAB              m1 (pt1,h1,f1)

                  J/Y       q
                                                                                These values are
                        Direction of J/Y                        P               decided by (q,P).
                                                      m2 (pt2,h2,f2)
                                                
        J / Y          f (q ) f ( p)  mm dqdp             q and P distributions are needed.

2007/09/03                                 B-physics meeting                                        12
Validation of di-muon trigger efficiency measurement

   • The result using trigger efficiency map is compared
     with that using L1_2MU06 flag.
   • For this study, use only J/Y reconstructed events.
   • Define di-muon trigger efficiency in two ways;
                         # of L1_2MU06 events
         1   ε flag                                      results
                      # of J/Y reconstruc events
                                                           ε flag  53.0%
         2   ε m ap 
                                   ε μ1ε μ2
                        # of J/Ψ reconstructed events      ε map  53.7%

         These two results are matched within 1%.
         -> Trigger efficiency map seems to work well.
         -> Error estimation should be done for the next step.
2007/09/03                            B-physics meeting                13
Background estimation

Fake J/Y ratio is estimated by counting the number of events
in 2.88GeV < Mmm < 3.3GeV.
                                                          From signal samples

         bbmu6X                                         Dijet samples
                              J/Y                                               J/Y
                                                        (all included)

             Fit only background

                                                             Main background is J1,J2
 Fake J/Y ratio
                                                             ※ More statistics are needed.
 bbmu6X:9.3%, Dijet:14.7% -> total 24.0%
2007/09/03                          B-physics meeting                                  14
Summary and outlook

• Single muon trigger efficiency is calculated with J/Y -> mm samples.
       ※ Identify pre-scaled trigger is needed for this study.
        The efficiency is well matched with that of single muon samples.
        The method seems to work well.
• Trigger efficiency map is made and di-muon trigger efficiency is
  calculated for J/Y reconstructed events.
        The result matches with that of using L1_2MU06 flag within 1%.
        The method seems to work well, too.
       ※ Samples including lower pt muons are needed for trigger efficiency map !!
• Background effects are estimated on bbmu6X and di-jet samples.
        For dijet samples, more statistics are needed.
• Calculate J/Y di-muon trigger efficiency.
• Estimate errors in di-muon trigger efficiency calculation.
• Use MU10 muon as seed muon and compare with the case of MU06.
  (trigger rate, statistics, etc…)

2007/09/03                             B-physics meeting                             15
Status of B-physics CSC note 7
B-physics CSC note 7

Contents of the note
• Introduction
• Method for obtaining trigger efficiency from data
• LVL1
       – Single muon efficiency at low-pT
       – Di-muon efficiency
• LVL2
       – Mufast
       – TrigDiMuon
• Proposal of calibration triggers
       – Calibration triggers necessary for MU6, 2MU6, MU6+X (basically, a
         prescaled single muon trigger)
       – J/psi signal events and trigger rate by the calibration trigger with respect to
• Conclusion

2007/09/03                               B-physics meeting                                 17
Status of the note

• Writing of the note has not started yet
       – Twiki page: CSCBPhysicsTask7

• Analysis
       – LVL1 study in a good shape
             • Demonstration of the method single muon efficiency done (comparison with
               single muon sample)
             • Expect to have the comparison of the obtained di-muon efficiency with truth soon
       – LVL2 study in a good shape for 2MU6, but the study just starting for
             • Need reconstruction by release 13 for TrigDiMuon
       – Consideration on the choice of thresholds and prescaled for the single muon
         trigger for calibration

• Plan to finish by the end of the year

2007/09/03                                  B-physics meeting                                 18