RPC and DTBX Ghosts and Track Quality

As mentioned before both trigger systems can artificially produce more tracks as output than has been fed into, which can lead to a higher or wrong trigger rate. In the single muon case this is a minor problem, which on the other hand aggravates with our requirement to trigger at most to four muons. From table 3 we see, that in the DTBX system this problem is much more prominent than for RPC. The reasons are due to a high extent left-right ambiguities of the Drift Tubes and ambiguous possibilities to reconstruct tracks. The sometimes complicated situation on the stage of ``trigger primitives'' which are the input to the regional muon trigger is still under investigation, as in this stage 50% ambiguities are the starting point, which fortunately are greatly reduced by the DTBX algorithms.

Analysing events, where ghost tracks appear, we recognize some common features :

• tracks of such events have mostly the same reconstructed and values
• such events appear in certain regions (see fig. 14)
  for RPC at
  for DTBX in complementary regions
• such events have ``lower quality'' tracks
• one of the two triggered muons has always a lower than

The appearance of ghost tracks is almost always connected to an insufficient number of hit detector elements. So in the RPC case (where at least 3 planes are needed for muon ``reconstruction''):

• 66% of events with ghost tracks have
  track 1 with 4 RPC planes used
  track 2 with 3 RPC planes used
• 33% of events with ghost tracks have
  track 1 and track 2 both with 3 RPC planes used
• is either 100 GeV/c or < 10 GeV/c

In the DTBX case, where two stations are the minimum requirement for ``reconstruction'':

• 77% of events with ghost tracks have both tracks with only two stations used, with each track having none station in common with the other track
• the most prominent configuration is where stations 1/2 and 3/4 are used
• mostly the configuration 3/4 will give a very wrong assignment

This leads to the requirement to introduce carefully designed ``quality bits'' to account quickly for these findings.

  
Figure 14: of events with ghost tracks for DTBX and RPC.

Similar effects can be seen in events with no ghost tracks found by the DTBX trigger. If only the two outer muon stations 3/4 are used to reconstruct a track, there is a high probability that the assigned is NOT reliable.