Reconstruction Tasks and Topics for Further Development

From GlueXWiki
Revision as of 14:46, 21 September 2011 by Mashephe (Talk | contribs) (Questions to Answer)

Jump to: navigation, search

Focus Areas

We will try to broadly classify outstanding software issues and tasks into three focus areas: Tracking, Calorimetry, and PID. For each of these areas we have appointed a contact person: Simon (Tracking), Matt (Calorimetry), and Paul (PID). The contact person is responsible for both prioritizing the issues to work on, seeing that progress is made on resolving these problems, and reporting status at each software meeting.


Tracking

Issues to Resolve

  1. wrong track sign choice at pattern recognition stage in FDC
  2. poor proton reconstruction in b1π events
    1. ...in general, need to improve tracking efficiency for multi-particle events...
  3. incorrect χ2 distributions from track fits: huge peak at zero
  4. validate propagation of errors from hit level through the fitting procedure to the track level

Questions to Answer

  1. How do we handle large single Coulomb scattering in tracking, e.g. between the CDC and FDC?
  2. How do we handle decays in tracking? (This might have similar resolution to the question above since the track shape is similar.)
  3. Can we quantify the performance of the track finding algorithm in a variety of event environments?

Calorimetry

Issues to Resolve

  1. allow easy switch between 1-2-3-4 and 3-3-4 BCAL schemes; include calibrations for both
  2. tune new BCAL reconstruction algorithm, explore impact on shower multiplicity in b1π events
  3. include timing reconstruction in FCAL
  4. correct the error matrix calculation for both FCAL and BCAL

Questions to Answer

  1. When photons get split into multiple clusters, is all of the energy detected?
  2. What is our efficiency for π0 detection? η→γγ?


PID

Tasks/Issues to Resolve

  1. Develop neutron reconstruction: calculate neutron energy from energy in calorimeters.
  2. Improve TOF/BCAL/FCAL uncertainties and integrate into PID.
  3. Implement TOF dE/dx calculation and integrate into PID.
  4. Look into scheme for timing PID: project RF time to timing detector or vice versa, look into determining if tracks in the same event from the hit times in the drift chamber
  5. Look into implementing BCAL dE/dx calculation.
  6. Look into weighting chi-squares from different sources so that they have equal weight.
  7. Determine required time resolution of start counter: does it need to be used to pick the RF beam bunch?
  8. Develop convention for defining the hit spacetime vertex for each detector.