(r20) NotesInPreparation < CMS

CMS Web>TrackerSoftware >NotesInPreparation (revision 20) (raw view)~~EditAttach~~
%TOC%

---+++Analysis notes

| *Author* | *Title* | *Files* | *Comment* |
| Ferenc Siklér | Reconstruction of low p<sub>T</sub> charged particles with the pixel detector | [[%ATTACHURL%/LowPtTracking.ps.gz][%ICON{ps}%]] [[%ATTACHURL%/LowPtTracking.pdf][%ICON{pdf}%]] | draft frozen |
| Ferenc Siklér | Reconstruction of V0s and photon conversions with the pixel detector | [[%ATTACHURL%/VzeroReconstruction.ps.gz][%ICON{ps}%]] [[%ATTACHURL%/VzeroReconstruction.pdf][%ICON{pdf}%]] | draft frozen |

---++++Changes made after the Physics Meeting

The modifications are listed here: [[NotesChanges][%ICON{html}%]]

#CmsSw
---+++Software in CMSSW

---++++ !Low p<sub>T</sub> tracking

The sources are available under =~sikler/public/CMSSW_2009-09-19/src=, the relevant changes are in this tgz file [[%ATTACHURL%/src.tgz][%ICON{zip}%]]. The resulted pixel tracks are in =/afs/kfki.hu/scratch/sikler/cmssw/tracker/pixeltracks.root=. 

   * *Changes to !RecoTracker/TkMSParametrization* %BR%
      !MultipleScatteringX0Data.{cc,h} is modified, reads in the rootified version of ORCA multiple scattering parametrisation, for the pixels only, at =~sikler/public/data/MultipleScatteringX0Data_pixel.root=.

   * *Changes to !RecoTracker/TkTrackingRegions/interface* %BR%
      Default is set to ptMin = 0.075 !GeV/c in !GlobalTrackingRegion.h. It should be set from configuration, as well as originRadius.

   * *Changes to !RecoPixelVertexing/PixelTriplets* %BR%
      _Modified triplet generator_:
         Classes !HitTripletGeneratorFromPairAndLayersWithHelix and !ThirdHitRZPredictionWithHelix provide the modified triplet finding algorithm. !PixelHitTripletGenerator.cc is modified in two places. It now asks for the evaluation of cluster info for all the reconstructed hits. At init() the choice of generator should be optional (propagated down from configuration), at the moment it is hardwired to the modified method. %BR%
      _Cluster shape:_
         Classes !ClusterData, !ClusterShape and !ClusterInfo provide compatibility check of the directed envelope of the cluster with the predicted movement of the track on the surface of the layer. !ClusterInfo reads in tables containing information on limits from =~sikler/public/data/{barrel,forward}.dat=, in text format at the moment. The analyzer !ClusterAnalyzer extracts these tables from comparison of simulated and reconstructed data, a configuration file clusteranalyzer.cfg is provided, !BuildFile changed accordingly. Hits can be locked in order to reduce fake rate for secondaries (second turn with wider originRadius).

   * *Addition to !RecoPixelVertexing/PixelTrackFitting/test* %BR%
      A small analyzer !LowPtTrackAnalyzer.cc plots simulated and reconstructed hits with tracks. A configuration file lowpttrackanalyzer.cfg is provided, !BuildFile changed accordingly.

   * To add %BR%
      Ambiguity resolver: if two tracks have at least two common hits, they should be merged or one of them killed. %BR%

   * Timing %BR%
      Hundred minimum bias Pythia events, starting from reconstructed hits, pixel triplet finding + track fitting. Time is without initialisation. Note that these are without ambiguity resolver.
    | *ptMin* | *Reconstructed tracks* | *Total time* | *Time/track* |
    | 0.075 !GeV/c | 1438 | 37 s | 25 ms |

---++++ !V0 finding

<p style="text-align: justify;">
The new packages !DataFormats/VZero and !RecoPixelVZero/PixelVZeroFinding are in cvs.
This is a fast V0- and conversion finder using pixel tracks. Actually it works with reco::Tracks provided all the relevant fields are filled.
</p>

1000 special events have been produced which contain 10 K<sub>S</Sub><sup>0</sup>, 10 &Lambda; and 10 &gamma;, all of them in the ranges -2<&eta;<2 and 2 !GeV/c<p<sub>T</sub><4 !GeV/c. The events with !PixelTracks are available at =/afs/kfki.hu/scratch/sikler/cmssw/pixeltracks.root=.

<p style="text-align: justify;">
Note that due to the default setting of !GlobalTrackingRegion (float ptMin = 1., float originRadius = 0.2) the track reconstruction is limited. This is why no dE/dx calculation was included for the identification of protons from &Lambda;s: it would not have helped at higher momenta. Of course all that can be substantially enhanced with low p<sub>T</sub> tracking. An associator class is not included yet.
</p>

   * Control plots:
      * root -q -b -l plot.C
      * Look at the produced [[%ATTACHURL%/test.ps][test.ps]]

---++++Simulation

[[%ATTACHURL%/NewTrackAction.cc][NewTrackAction.cc]]

---+++Software in ORCA

Create project area ORCA_8_13_0 or later.
<blockquote><verbatim>
scramv1 project ORCA ORCA_8_13_x
cd ORCA_8_13_x/src
</verbatim></blockquote>

Check out (=cvs co=) or copy (=cp -r=) subsystems !TrackerReco/PixelTrackFinder, !TrackerReco/PixelTrackFit and !TrackerReco/TkHitTriplets. The relevant changes are in this tgz file [[%ATTACHURL%/TrackerReco.tgz][%ICON{zip}%]]. Overwrite and add new files by =tar -xvzf !TrackerReco.tgz=.
Compile with =eval `scramv1 runtime -csh`; scramv1 build=. A new subsystem !TrackerReco/LowPt was created, go to !TrackerReco/LowPt/test. 

The file orcarc.aa contains parameters. The tunable ones are:
<blockquote><verbatim>
LowPt:ptMin             = 0.175
LowPt:originRadius      = 0.1
LowPt:originHalfLength  = 0.1
</verbatim></blockquote>

Enter proper !InputCollections and !InputFileCatalogURL values. Start reconstruction by =TestLowPt -c orcarc.aa=.
A sample area with data files is set up on AFS below =~sikler/public/ORCA_8_13_0=.

---+++Other notes in preparation

| *Author* | *Title* | *Files* | *Comment* |
| Ferenc Siklér | Particle identification using energy loss in the pixel detector |  [[%ATTACHURL%/energyloss.ps.gz][%ICON{ps}%]] [[%ATTACHURL%/energyloss.pdf][%ICON{pdf}%]] | first draft |
| Gergely Patay | Phi meson reconstruction using low p<sub>T</sub> tracking and dE/dx in the pixel detector |  [[%ATTACHURL%/phinote.pdf][%ICON{pdf}%]] | 0th draft |

<!--
   * Set ALLOWTOPICCHANGE = Main.CmsHiGroup
   * Set ALLOWTOPICRENAME = Main.CmsHiGroup
   * Set NOSEARCHALL = on
--> 

-- Main.FerencSikler - 20 Jun 2006
Attachments
Topic attachments
I	Attachment	History	Action	Size	Date	Who	Comment
tgz	src.tgz	r3 r2 r1	manage	14.0 K	2006-09-21 - 14:30	FerencSikler	Low pT tracking for CMSSW_2009-09-19
ps	test.ps	r2 r1	manage	29.9 K	2006-08-23 - 05:37	FerencSikler
Topic revision: r20 - 2006-11-14 - FerencSikler
CMS (CERN)
Copyright &Â© by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback