Mattione Update 09072011

Summary

• OBJECTIVE: I'm updating the reconstruction uncertainties to improve the particle-ID algorithms and support the kinematic fitting analyses. Right now the tracking uncertainties are yielding dubious chi-squares and kinematic fits with too many events with low confidence levels. Also, the timing uncertainties used for calculating the particle-id FOM are one-size-fits-all numbers, instead of incorporating any momentum or energy dependence.

• PLAN: I'm comparing the generated and reconstructed track parameters in the TOF, BCAL, and FCAL systems to determine the uncertainties.

• STATUS: I've almost finished getting the comparisons, selected results are shown below. There is a bug in the FCAL time reconstruction, and the FCAL energy uncertainty is significantly underestimated.

FCAL Time Reconstruction

• The reconstructed FCAL time is significantly offset from the simulated times.

• hd_dump of DFCALShower, DFCALTruthShower, and DFCALHit. Note that both the DFCALShower and DFCALHit times are ~2+ ns greater than the DFCALTruthShower times.

Event: 1
DFCALShower:
 E(GeV):   X(cm):   Y(cm):   Z(cm):   t(ns):
--------------------------------------------
   3.27    -8.27    63.70   642.36    21.16  //Photon generated at target center (0, 0, 65), so path length = 580.922 and t should be = 19.377

DFCALTruthShower:
 x(cm): y(cm): z(cm): px(MeV/c): py(MeV/c): pz(MeV/c):   E(MeV): t(ns): primary: track: type:
---------------------------------------------------------------------------------------------
  -8.0   61.7  625.3      -44.8      343.8     3123.6  3142.732   18.8        1      1     1  //Photon generated at target center (0, 0, 65), so path length = 563.744 and t should be = 18.8
//similar, additional hits were present but were truncated

DFCALHit:
  row: column: x(cm): y(cm):   E(MeV):  t(ns):
----------------------------------------------
   44      26  -12.0   60.0    42.050  21.082 
//similar, additional hits were present but were truncated

• In programs/Simulation/HDGeant/hitFCal.c the times saved to the s_FcalTruthShowers_t and s_FcalTruthHit_t objects are different: the s_FcalTruthHit_t times (which are later smeared and converted to DFCALHit objects) are increased to account for what looks like light propagation to the PMTs (line ~192). Perhaps this isn't being taken into account in the DFCALCluster reconstruction code??? I'm not really familiar with it, so I'm hoping that someone who is will take a look at this.

FCAL Energy Uncertainty

• Difference between reconstructed and thrown photon energy, vs. photon energy.

• Next, the above histogram was fit to Gaussian functions, and the uncertainties are plotted with the points below. The function shows the FCAL shower energy uncertainty as currently used in the reconstruction code.

TOF Time Uncertainty

• Time Uncertainty (in reconstruction code, the particle time uncertainty at the vertex is set to 80 ps).