1 / 24

Stefano Spataro

Event Reconstruction in the PandaRoot Framework. Stefano Spataro. for the collaboration. ISTITUTO NAZIONALE DI FISICA NUCLEARE Sezione di Torino. Tuesday, 22 th May, 2012. Overview. The Panda Experiment The FairRoot and PandaRoot frameworks Tracking

moral
Télécharger la présentation

Stefano Spataro

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Event Reconstruction in thePandaRootFramework Stefano Spataro for the collaboration ISTITUTO NAZIONALE DI FISICA NUCLEARE Sezione di Torino Tuesday, 22th May, 2012

  2. Overview • The Panda Experiment • The FairRoot and PandaRoot frameworks • Tracking • Particle Identification • Analysis • Time Based Simulation

  3. The Panda experiment pp, pA collisions 1.515 GeV/c (p momentum) AntiProton Annihilations at Darmstadt Multi purpose detector at FAIR Darmstadt (Germany) GSI • Charmonium (cc) spectroscopy • Open charm spectroscopy • Search for gluonic excitations (hybrids - glueballs) • Charmed hadrons in nuclei • Drell-Yan • Single and double Hypernuclei • Parton Dist., EM Form Factor… Facility for Antiproton and Ion Research

  4. G3VMC Geant3 Geometry Virtual MC G4VMC Geant4 Root files Hits, Digits, Tracks Cuts, processes Application IO Manager Track propagation Run Manager RTDataBase Event Display Event Generator Magnetic Field Detector base Tasks Hit Producers STT Dipole Map DPM TOF Solenoid Map EMC digitizers EvtGen MUO const. field MVD Track finding ASCII DIRC Panda Code FTS Pythia GEM Code Design FairRoot Root files M.Al-Turany, D.Bertini, F.Uhlig, R.Karabowicz Oracle MySQL TSQLServer Postgresql PandaRoot CbmRoot R3BRoot MPDRoot (NICA) ASYEOSRoot EICRoot

  5. The PandaRoot framework Virtual Monte-Carlo (2.13) ROOT (5.32) based on dynamic data structure (based on ROOT Trees and Folders) use of many ROOT application (TGeo, EVE, TMVA, PROOF, TSQLServer) same geometry/code for Geant3 - Geant4 (9.5) compiled and running on more than 10 Linux platforms + Mac OS X Alien2 based GRID Tracking TDR data production collaboration with external developers CERN/ALICE FAIR/HADES-CBM-R3B

  6. Global Tracking p FTS MVD GEM STT

  7. Kalman Filter (GENFIT- Munich*) Detector Hits Prefit Track Tracking: Global Fit Energy loss Not homogeneous magnetic field Different detector hits • 3D points – TPC • planar hits – MVD/GEM • tube + drift time – STT/FTS barrel forward Track Follower (GEANE – Pavia**) **A.Fontana, L.Lavezzi, A.Rotondi same geometry for simulation and track following *F.Böhmer, C. Höppner, S.Neubert, J.Rauch  Poster: New Developments in the GENFIT track fitting framework

  8. Barrel Tracking: Pattern Recognition 1° step – MVD/STT local pattern recognition STT MVD 2° step – Correlation of STT & MVD tracklets - Correlation with STT/MVD spurious hits 3° step – Extrapolation to GEM planes 4° step – Kalman Filter G.Boca, R. Karabowicz , L.Lavezzi, T.Stockmanns

  9. Barrel Tracking: Performances STT stand-alone STT+MVD+GEM Large Improvement from MVD/GEM detectors S.Costanza, L.Lavezzi

  10. _ p @ [GeV/c] 15.00 11.91 8.90 4.06 1.50 Forward Tracking FTS + MVD + GEM • Ideal Pattern Recognition • Kalman Filter muons |B| [T] E.Fioravanti, I.Garzia, R.Kliemt

  11. Particle Identification Forward EMC EMC p FWDMDT TOF DIRC MDT FTOF

  12. Particle Identification Implemented PDFs for many detectors (Bayes) EMC STT MVD DIRC DISC MDT A.Cecchi, L.Lavezzi, R.Kunne, Y.Lang, L.Zotti

  13. Interface to Analysis Flexible way to switch between different algorythms PndPidCandidate PndPidProbability • PidChargedCand • Momentum • STT dE/dx • EMC energy • EMC shower shape • Cherenkov angle • MVD dE/dx • # Muon Layers • … • PidNeutralCand • Momentum • STT dE/dx • EMC energy • EMC shower shape • Cherenkov angle • MVD dE/dx • # Muon Layers • … • MVD • P(e) • P() • P() • P(K) • P(p) • … • EMC SS • P(e) • P() • P() • P(K) • P(p) • … • DIRC • P(e) • P() • P() • P(K) • P(p) • … • … • P(e) • P() • P() • P(K) • P(p) • … PndAnaPidCombiner analysis->FillList(pip, “PionLoose”,”BEMC;BMVD;BDIRC”) selection criterium candidate list algorithms to merge K. Götzen, R.Kliemt

  14. MultiVariate Particle Identification implementation of TMVA methods • EMC shower shape analysis e/ separation in EMC preliminary tests for muons (MDT) M.Babai. G.Gumberidze, D.Melnichuk  Correlation Variables 

  15. reconstructed refitted Rho Toolkit Analysis ’ 4C constraints • Implementation of • kinematic fitters • vertex fitters • fast simulation reconstructed refitted K.Götzen, R.Kliemt, V.Jha

  16. Reconstruction of Physics Channels – D+D- Heavy-Light Systems kaons pions • 6 charged tracks in the final state • Secondary vertexing

  17. M.Mertens

  18. D-Meson Vertex Resolution M.Mertens

  19. Traditional DAQ

  20. self-triggered detector front-end Panda DAQ continuous sampling DAQ flash ADC each signal get a time stamp (155.52 MHz) high quality clock distributed front-end feature extraction (signal amplitude, shape, …)

  21. Time Dependent Simulation slow transition from event based to time ordered simulation

  22. STT MIXING physics event background events background events t1 t2 t3 t4 t0 t5 t6 t7 t8 t9 time 100 nsec window 100 nsec window w/o Event Mixing w/ Event Mixing @ 20 MHz interaction rate G.Boca

  23. x MVD Points • STT Points Time Based Simulation • Randomized Digi Data • Sorting Digi Data using Time Stamps + Drift Time, ToT… Time Stamp [ns] Time Stamp [ns] MVD Digi Data Stream same color = same event Position in Array Digi Index Digi index M.Al-Turany, T.Stockmanns

  24. Summary • PandaRoot is the framework for the Panda full simulation • Used successfully for massive data production on PandaGrid • Reconstruction • CentralTracking in the Barrel complete (submitted TDRs for STT&MVD) • Forward Tracking  just started, promising results • Bayesian Particle Identification available, flexibility at the analysis stage • Investigation on TMVA methods for PID • Analysis of physics benchmark channels  ready ! • Time Based Simulation • New concept, challenging online and offline reconstruction • Basic Event Mixing algorithms and cleaning code ready • Time Based structure implemented in the framework and in some detectors • Redesign reconstruction code…

More Related