Reference Quantities

For simulated data the following reference quantities are selected:

1. for MC:
   • reconstruction efficiencies of reference tracks,
   • clone and ghost rate,
   • residuals of track parameters ($x,y,tx,ty$),
   • number of hits used for the reconstruction of a track,
   • number of unused hits in the event (hits which are not used for the reconstruction of any tracks),
   • percentage of tracks that can be correctly tracked through the TC area,
   • percentage of correctly matched VDS-PC segments,
   • time consumption.
2. for data:
   • number of hits used for the reconstruction of a track,
   • number of unused hits in the event,
   • percentage of tracks that can be tracked through the TC area,
   • percentage of matched VDS-PC segments,
   • time consumption,
   • agreement with MC.

For evaluation purposes all simulated and reconstructed tracks are subdivided into several categories: a ``reference set'' of tracks, an ``all set'', an ``extra set'', clone and ghost tracks.

By definition, a track from the all set of tracks should intersect the sensitive regions of at least 5 layers (roughly 1 superlayer). A reference track should have a momentum greater than 1 GeV/c and intersect the sensitive regions of more than 17 layers (this number corresponds roughly to 3 traversed superlayers). Note, that the requirement of more than 17 traversed layers is simply an acceptance cut and does not imply the presence of more than 17 hits, since hit inefficiencies and track overlaps might reduce the number of hits considerably. Therefore, the obtained numbers for the reconstruction efficiency depends on the hit efficiencies of the detectors. All track sets are summarized below:

• reference tracks ( particle momenta larger than 1 GeV/c, more than 17 layers in the PC area are crossed),
• all tracks, (more than 5 layers are crossed in the PC area),
• physics tracks,
  • primary tracks,
    • all primaries, more than 5 layers are crossed in the PC area,
    • reference primaries, more than 17 layers are crossed in the PC area and particle momenta larger than 1 GeV/c,

  • lepton tracks from $J/\psi$ decays,
    • all $J/\psi$-leptons, more than 5 layers are crossed in the PC area,
    • reference $J/\psi$-leptons, more than 17 layers are crossed in the PC area and particle momentum large than 5 GeV/c.

In addition to these tracks the so-called extra set of tracks, containing short high-momentum and low-momentum tracks, was also considered.

A reconstructed track is assigned to a generated particle, if at least 70% (70% criteria [#!MAN!#]) of its hits have been caused by this particle. A generated particle is regarded as found, if it has been assigned to at least one reconstructed track. Ghost tracks are reconstructed tracks for which no simulated tracks were found. The ghost rate is normalized to all reconstructed tracks. Clones are reconstructed tracks for which a simulated track is found, but a better match exists (If one simulated track causes n tracks to be reconstructed, the match with the highest number of hits agreeing is called ``the right track'', the n-1 other tracks are counted as clones. If the number of agreeing hits is equal, the track with better parameters is ``the right track''). The clone rate is normalized to all reconstructed tracks.