GSI Forum
GSI Helmholtzzentrum für Schwerionenforschung

Home » PANDA » PandaRoot » Tracking » Cleanup functions
Cleanup functions [message #17360] Tue, 30 September 2014 20:03 Go to next message
Gianluigi Boca is currently offline  Gianluigi Boca
Messages: 177
Registered: March 2004
first-grade participant
From: *pv.infn.it
dear collaborators,
I put in svn a new version of
tracking/PndTrkTracking2.cxx end .h

containing a new feature which is the possibilty of activating the Cleanup code (code necessary to remove ghost tracks produced
in the Pattern Recognition by the 20 MHz pileup events).

BY DEFAULT THIS CLEANUP IS DEACTIVATED.

But if, in his reconstruction Macro, after :

PndTrkTracking2* tracking = new PndTrkTracking2(0,true,true,true);

the user put the following line :

tracking->CleanupMvd();

then it activates the Cleanup procedure based on the Mvd hits present in the tracks;
on the other hand if

tracking->CleanupStt();

then it activates the Cleanup procedure based on the Stt hits present in the tracks;
and finally if

tracking->Cleanup();

it activates both.

FOR THE TIME BEING IN THE ANALYSIS OF IDEAL EVENTS (no pileup due to 20 MHz interaction rate)
I suggest not to use any of this Cleanups.
On the contrary, when analysing the more realistic events with pileup the Cleanups are necessary.

For the time being the Cleanup procedures tend to overkill the problem and the reconstruction
efficiency is in general lower than normal.

I AM WORKING RIGHT NOW IN THE REFINEMENT OF THE CLEANUP PROCEDURES and I hope to arrive soon at a good
situation.

Cheers

Gianluigi



Re: Cleanup functions [message #17367 is a reply to message #17360] Thu, 02 October 2014 15:40 Go to previous messageGo to next message
StefanoSpataro is currently offline  StefanoSpataro
Messages: 2736
Registered: June 2005
Location: Torino
first-grade participant

From: *to.infn.it
Dear all,
I tried the last tracking code, switch all the possible cleanup options, and verifying the analysis results from the standard nechmark channel (psi(2s) -> J/psi pi+ pi-) without any background mixing.

Here you can see the j/psi and the psi(2s) invariant mass distributions:

index.php?t=getfile&id=8120&private=0

At a first glance, it seems the cleanup code affects mostly low p tracks (the pions), while the efficiency values for the high p (muons from j/psi) are consistent. Moreover, Mvd cleanup does not reduce the counts so much, the strongest effect comes from the stt cleanup. Of course the code should be verified with the event mixing (thing that I am NOT going to do now).
  • Attachment: cucomp.gif
    (Size: 17.00KB, Downloaded 499 times)
Re: Cleanup functions [message #17368 is a reply to message #17367] Fri, 03 October 2014 16:25 Go to previous message
Gianluigi Boca is currently offline  Gianluigi Boca
Messages: 177
Registered: March 2004
first-grade participant
From: *pv.infn.it
hi Stefano,
thank you for showing those results.
I stress again that the Cleanup functions are not the final ones since I am still working on them

Gianluigi


StefanoSpataro wrote on Thu, 02 October 2014 15:40
Dear all,
I tried the last tracking code, switch all the possible cleanup options, and verifying the analysis results from the standard nechmark channel (psi(2s) -> J/psi pi+ pi-) without any background mixing.

Here you can see the j/psi and the psi(2s) invariant mass distributions:

index.php?t=getfile&id=8120&private=0

At a first glance, it seems the cleanup code affects mostly low p tracks (the pions), while the efficiency values for the high p (muons from j/psi) are consistent. Moreover, Mvd cleanup does not reduce the counts so much, the strongest effect comes from the stt cleanup. Of course the code should be verified with the event mixing (thing that I am NOT going to do now).

Previous Topic: double tracks in offline Pattern Recognition
Next Topic: antiprotons lost above 10 degrees
Goto Forum:
  


Current Time: Sat Dec 14 08:04:36 CET 2024

Total time taken to generate the page: 0.00772 seconds