GSI Forum
GSI Helmholtzzentrum für Schwerionenforschung

Home » PANDA » PandaRoot » Tracking » track finder
track finder [message #5775] Thu, 31 January 2008 16:02 Go to next message
asanchez is currently offline  asanchez
Messages: 350
Registered: March 2006
first-grade participant
From: *kph.uni-mainz.de
Dear all,
i have a (maybe very stupid) question Rolling Eyes
concerning the track finder algorithm
in both detectors tpc(rieman or genfit) and geane(stt).

Is the track finder considering
only primaries tracks or is considering
also the secondaries(decays)tracks?

thanks in advance

ALicia S.
Re: track finder [message #5776 is a reply to message #5775] Thu, 31 January 2008 16:12 Go to previous messageGo to next message
Sebastian Neubert is currently offline  Sebastian Neubert
Messages: 282
Registered: March 2006
Location: Munich
first-grade participant

From: *e18.physik.tu-muenchen.de
Hi Alicia!

The riemann track finder should find all tracks. There is no restriction to primary tracks!

Cheers! Sebastian.


Sebastian Neubert
Technische Universität München
Department Physik E18
sneubert@e18.physik.tu-muenchen.de
tel: +49-8928912592
Re: track finder [message #5777 is a reply to message #5775] Thu, 31 January 2008 17:10 Go to previous messageGo to next message
Pablo Genova is currently offline  Pablo Genova
Messages: 32
Registered: May 2007
continuous participant
From: *pv.infn.it
Hi Alicia,

also the stt track finder is not restricted to primary tracks, as far as I know.
However take into account that the pattern recognition is ideal, i. e. we are still using the Monte Carlo truth.

Also geane can be used both for primaries and for secondaries.

ciao, Pablo
Re: track finder [message #5811 is a reply to message #5775] Tue, 05 February 2008 11:08 Go to previous messageGo to next message
asanchez is currently offline  asanchez
Messages: 350
Registered: March 2006
first-grade participant
From: *gsi.de
Dear all,
i have another question concerning the
kalman fit implemented for tpc(genfit )
and in stt(geane + kalman?).
Additionally to the fit parameters
such as momentum, dip angle,...
Is it possible to access to the arc length(s)
of the track (XY proyection)
after the fit with the kalman algorithm?

If yes, i would to know how.

thanks in advance
Alicia.
Re: track finder [message #5827 is a reply to message #5775] Wed, 06 February 2008 12:44 Go to previous messageGo to next message
Lia Lavezzi
Messages: 291
Registered: May 2007
Location: Torino
first-grade participant

From: *pv.infn.it
Hi Alicia,
concerning geane, in the case of propagation to closest approach to the wire/point, there is a function of CbmGeanePro called GetLengthAtPCA() which returns the track length at the point of closest approach. This returns the complete track length (not only the xy projection).

Concerning the propagation to a defined volume or plane this function does not work, since the variable trklength is not filled.

What do you need exactly?

Ciao,
Lia.
Re: track finder [message #5834 is a reply to message #5827] Fri, 08 February 2008 10:33 Go to previous messageGo to next message
asanchez is currently offline  asanchez
Messages: 350
Registered: March 2006
first-grade participant
From: *kph.uni-mainz.de
Hi Lia taking a look into the tracks parameters
providing by the kalman filter,

one has access to
P momentum
Lambda dip angle
Qp charge/momentun

X0
Y0 after circle fit

Z0 after fitting the line arclength(XY plane)vs Z

Iis it true? or I'm wrong. If not
could you tell me which are the parameters gived by the kalman fit
and in which coordinate system ?

thank you very much
ALiciaS.
Re: track finder [message #5835 is a reply to message #5834] Fri, 08 February 2008 11:37 Go to previous messageGo to next message
Lia Lavezzi
Messages: 291
Registered: May 2007
Location: Torino
first-grade participant

From: *pv.infn.it
Hi Alicia,
after the Kalman fit you have the track in the SD frame representation, so the 5 parameters are: q/p, v', w', v and w. You can access them with the functions in CbmTrackParP and CbmTrackPar (from which it inherits): GetQp(), GetV(), and so on. You can also access the x, y, z coordinates and the px, py, pz momentum components in the master reference system (you find the functions in the same classes).

Since you talk about the dip angle, and maybe you need it, I explain you how to get it: you need to change your representation, from the SD system (CbmTrackParP) to the SC one (CbmTrackParH), where the accessible parameters are q/p, lambda, phi, yperp, zperp. To do this you should use the transformation functions in CbmGeaneUtil. Consider as an example in the kalstt tutorial, in kalstt/GeaneEx/CbmGeaneTrKalStt.cxx, the lines from 298 to 327: they actually perform the reverse transformation, from SD to SC.

One more remark: the Kalman filter procedure does not divide the fit problem into two planes (x-y and z-track length), it is performed on a virtual detector plane; the separation of the problem in the two planes is used only during the prefit (that we perform with an helix in PndSttHelixTrackFitter).

Hope this helps, but if you need more info please ask Smile
Ciao,
Lia.



Re: track finder [message #5836 is a reply to message #5835] Fri, 08 February 2008 11:46 Go to previous messageGo to next message
asanchez is currently offline  asanchez
Messages: 350
Registered: March 2006
first-grade participant
From: *kph.uni-mainz.de
Hi Lia, thank you for your information Smile

I have more questions Razz

In the xyz cordinates system,
What are xPerp yPerp related to? Are they representing
some kind of Detector plane(alla Genfit)?

Is then the beam direction paralell to x-Axis?

thank you in advance. Surprised

ALiciaS.

Re: track finder [message #5837 is a reply to message #5836] Fri, 08 February 2008 12:25 Go to previous message
Lia Lavezzi
Messages: 291
Registered: May 2007
Location: Torino
first-grade participant

From: *pv.infn.it
Hi Alicia,
first let' s talk about the "perp" coordinates, so about the CbmTrackParH representation: the yperp and zperp (not xperp-yperp!) define a plane orthogonal to the particle direction. When you use the CbmTrackParH representation, you describe the track in the SC frame, which is defined by the three orthogonal axes xperp, yperp, zperp, where xperp is along the particle direction, yperp is perpendicular to xperp and parallel to the xy plane (in the master reference system, MARS) and zperp is the third axis chosen in order to have an orthonormal reference frame. So the yperp-zperp plane turns out to be a plane orthogonal to the particle direction.
So, concerning the second question, the beam direction is not chosen parallel to the x axis in MARS, it is generated random, but the xperp axis in SC is in each point chosen parallel to the particle direction.

Just one clarification: for the STT we use a different detector plane, which is not perpendicular to the track and that' s why we use the CbmTrackParP representation instead of the CbmTrackParH one. CbmTrackParP, let' s say, allows you to define your own detector plane (virtual or not) and to get the track parameters on it.

Ciao,
Lia.


Previous Topic: Weird STT reconstruction: efficiency depends on charge
Next Topic: riemann track finding
Goto Forum:
  


Current Time: Mon Oct 07 20:07:48 CEST 2024

Total time taken to generate the page: 0.00786 seconds