class AliITStrackerMI: public AliTracker

Default constructor

This is the AliITStrackerMI constructor

destructor

This function read ITS bad detectors, chips, channels from AliITSDetTypeRec
i.e. from OCDB

This function loads ITS clusters

This function unloads ITS clusters

 Publishes all pointers to clusters known to the tracker into the
 passed object array.
 The ownership is not transfered - the caller is not expected to delete
 the clusters.

 Correction for the material between the TPC and the ITS

 This functions reconstructs ITS tracks
 The clusters must be already loaded !

 This functions propagates reconstructed ITS tracks back
 The clusters must be loaded !

 This functions refits ITS tracks using the
 "inward propagated" TPC tracks
 The clusters must be loaded !

       Return pointer to a given cluster

 Get track space point with index i

 Get track space point with index i
 (assign error estimated during the tracking)

 Follow prolongation tree

 This function computes the rectangular road for this track

 Returns the thickness between the current layer and the vertex (units X0)

 Returns the array of layers between the current layer and the vertex

 This function refits the track "track" at the position "x" using
 the clusters from "clusters"
 If "extra"==kTRUE,
    the clusters from overlapped modules get attached to "track"
 If "planeff"==kTRUE,
    special approach for plane efficiency evaluation is applyed

 This function refits the track "track" at the position "x" using
 the clusters from array
 If "extra"==kTRUE,
    the clusters from overlapped modules get attached to "track"
 If "planeff"==kTRUE,
    special approach for plane efficiency evaluation is applyed

 calculate normalized chi2
  return NormalizedChi2(track,0);

 return matching chi2 between two tracks

  return probability that given point (characterized by z position and error)
  is in SPD dead zone
     This method assumes that fSPDdetzcentre is ordered from -z to +z

 calculate normalized chi2

 calculate normalized chi2
  if (forwardtrack->fNUsed>0.3*float(forwardtrack->GetNumberOfClusters())) return 10000;

       Return pointer to a given cluster

 register track to the list

 unregister track from the list

get number of shared clusters

 find first shared track

 mean  number of clusters

 try to find track hypothesys without conflicts
 with minimal chi2;

 This function marks clusters assigned to the track

 add track to the list of hypothesys

 compress array of track hypothesys
 keep only maxsize best hypothesys

 try to find best hypothesy
 currently - minimal chi2 of track+backpropagated track+matching to the tpc track
 RS: optionally changing this to product of Chi2MIP(0)*Chi2MIP(3) == (chi2*chi2_interpolated)

 get "best" hypothesys

 RS: flag those tracks which are suxpected to have fake clusters

This function "cooks" a track label. If label<0, this track is fake.

 Fill the dE/dx in this track

 Create some arrays

 Compute predicted chi2

 Take into account the mis-alignment (bring track to cluster plane)

 Update ITS track

DCA sigmas parameterization
to be paramterized using external parameters in future

 Clusters from delta electrons?

 Update ESD track

 Get nearest upper layer close to the point xr.
 rough approximation

 Fill a look-up table with mean material

 Propagate beyond beam pipe and correct for material
 (material budget in different ways according to fUseTGeo value)
 Add time if going outward (PropagateTo or PropagateToTGeo)

 Define budget mode:
 0: material from AliITSRecoParam (hard coded)
 1: material from TGeo in one step (on the fly)
 2: material from lut
 3: material from TGeo in one step (same for all hypotheses)

 Propagate beyond SPD or SDD shield and correct for material
 (material budget in different ways according to fUseTGeo value)
 Add time if going outward (PropagateTo or PropagateToTGeo)

 Define budget mode:
 0: material from AliITSRecoParam (hard coded)
 1: material from TGeo in steps of X cm (on the fly)
    X = AliITSRecoParam::GetStepSizeTGeo()
 2: material from lut
 3: material from TGeo in one step (same for all hypotheses)

 Propagate beyond layer and correct for material
 (material budget in different ways according to fUseTGeo value)
 Add time if going outward (PropagateTo or PropagateToTGeo)

 Define budget mode:
 0: material from AliITSRecoParam (hard coded)
 1: material from TGeo in stepsof X cm (on the fly)
    X = AliITSRecoParam::GetStepSizeTGeo()
 2: material from lut
 3: material from TGeo in one step (same for all hypotheses)

 Initialize LUT for storing material for each prolonged track

 Delete LUT for storing material for each prolonged track

 Check if we are forced to skip layers
 either we set to skip them in RecoParam
 or they were off during data-taking

 This method is used to decide whether to allow a prolongation
 without clusters, because we want to skip the layer.
 In this case the return value is > 0:
 return 1: the user requested to skip a layer
 return 2: track outside z acceptance

 This method is used to decide whether to allow a prolongation
 without clusters, because there is a dead zone in the road.
 In this case the return value is > 0:
 return 1: dead zone at z=0,+-7cm in SPD
     This method assumes that fSPDdetzcentre is ordered from -z to +z
 return 2: all road is "bad" (dead or noisy) from the OCDB
 return 3: at least a chip is "bad" (dead or noisy) from the OCDB
 return 4: at least a single channel is "bad" (dead or noisy) from the OCDB

 check dead zones at z=0,+-7cm in the SPD

 Gives position of track in local module ref. frame

 Method to be optimized further:
 Aim: decide whether a track can be used for PlaneEff evaluation
      the decision is taken based on the track quality at the layer under study
      no information on the clusters on this layer has to be used
      The criterium is to reject tracks at boundaries between basic block (e.g. SPD chip)
      the cut is done on number of sigmas from the boundaries

  Input: Actual track, layer [0,5] under study
  Output: none
  Return: kTRUE if this is a good track

 it will apply a pre-selection to obtain good quality tracks.
 Here also  you will have the possibility to put a control on the
 impact point of the track on the basic block, in order to exclude border regions
 this will be done by calling a proper method of the AliITSPlaneEff class.

 input: AliITStrackMI* track, ilayer= layer number [0,5]
 return: Bool_t   -> kTRUE if usable track, kFALSE if not usable.

 This Method has to be optimized! For the time-being it uses the same criteria
 as those used in the search of extra clusters for overlapping modules.

 Method Purpose: estabilish whether a track has produced a recpoint or not
                 in the layer under study (For Plane efficiency)

 inputs: AliITStrackMI* track  (pointer to a usable track)
 outputs: none
 side effects: update (by 1 count) the Plane Efficiency statistics of the basic block
               traversed by this very track. In details:
               - if a cluster can be associated to the track then call
                  AliITSPlaneEff::UpDatePlaneEff(key,kTRUE);
               - if not, the AliITSPlaneEff::UpDatePlaneEff(key,kFALSE) is called

 find the MC cluster for the label

 This function sets up flags and cuts for the first tracking pass

   flags[0] - vertex constaint flag
              negative means "skip the pass"
              0        means "no constraint"
              positive means "normal constraint"

 This function sets up flags and cuts for the second tracking pass

   flags[0] - vertex constaint flag
              negative means "skip the pass"
              0        means "no constraint"
              positive means "normal constraint"

This function "cooks" a track label. If label<0, this track is fake.

get normalize chi2

 methods for debugging (RS) >>