df/df7/examples_2tracking_2config_81b_2reconstruction_8C_source.html

//

// Track reconstruction: 1) no track finder, namely use "ideal" hit-to-track

// association, 2) run "real" Kalman filter (KF) fit on digitized hits;

//


void reconstruction()

{

  // Load basic libraries;

  gROOT->Macro("$VMCWORKDIR/gconfig/rootlogon.C");


  // Create generic analysis run manager; configure it for track reconstruction;

  EicRunAna *fRun = new EicRunAna();

  // Import the files with simulated events and digitized hits;

  fRun->SetInputFile ("simulation.root");

  fRun->AddFriend    ("digitization.root");

  // Define the output file with the reconstructed tracks;

  fRun->SetOutputFile("reconstruction.root");


  // Call "ideal" hit-to-track associator ("track finder") routine;

  EicIdealTrackingCode* idealTracker = new EicIdealTrackingCode();

  // Default in this example is to use "FWDST" detector hits; more than

  // one custom detector may be added here via AddDetectorGroup() call(s);

  idealTracker->AddDetectorGroup("IP");

  idealTracker->AddDetectorGroup("FWDST");

  // Add a bit of fairness to the reconstruction procedure; smear "ideal"

  // momenta by 10% (relative) before passing the hit collection over to the KF fitter;

  idealTracker->SetRelativeMomentumSmearing(0.1);

  // Also smear a bit the "ideal" vertex;

  idealTracker->SetVertexSmearing(100 * eic::um, 100 * eic::um, 100 * eic::um);

  // Assume 100% hit association efficiency;

  idealTracker->SetTrackingEfficiency(1.0);

  fRun->AddTask(idealTracker);


  // Invoke and configure PandaRoot Kalman filter code wrapper; use hit-to-track

  // association of the detectors included via idealTracker->AddDetectorGroup() calls

  // during the "ideal track finder" call above;

  EicRecoKalmanTask *kftask = new EicRecoKalmanTask(idealTracker);

  // Store Monte-Carlo (truth) and reconstructed track parameterizations at

  // every detector plane, where a particular track produced hits; if other

  // locations along the track are of interest, thin additional "profiling" planes

  // need to be adde to the setup; one can consider to provide interpolation and

  // extrapolation tools right in the analysis.C script, but as of now (02/04/2020)

  // this feature is not implemented;

  kftask->StoreTrackParameterization();

  fRun->AddTask(kftask);


  // Perform track backward propagation to the beam line;

  fRun->AddTask(new PndPidCorrelator());


  // Initialize and run the reconstruction; exit at the end;

  fRun->Run();

} // reconstruction()