ILC Detector Software Development

1
ILC Detector Software Development

• Norman Graf
• ALCPG Physics Detector Simulation
  Reconstruction
• US ILC Detector RD Program Review
• ANL, June 20, 2007

2
Software Development Overview

• Physics analyses.
• Refine and strengthen the arguments for the ILC.
• Detector design.
• Integrated system design for an optimal detector.
• Event reconstruction.
• Demonstrate that proposed detector systems can
  conduct the physics program, and allows
  cost-benefit optimization to be done.
• Testbeam and prototype infrastructure.
• Support subsystem design, readout and analysis.

3
LCD Simulation Mission Statement

• Provide full simulation capabilities for Linear
  Collider physics program
• Physics simulations
• Detector designs
• Reconstruction and analysis
• Need flexibility for
• New detector geometries/technologies
• Different reconstruction algorithms
• Limited resources demand efficient solutions,
  focused effort.
• Strong connections between university groups,
  national labs, international colleagues.

4
Goals

• Facilitate contribution from physicists in
  different locations with various amounts of
  available time.
• Use standard data formats, collaborate
  cooperate.
• Provide a general-purpose framework for physics
  software development.
• Develop a suite of reconstruction and analysis
  algorithms and sample codes.
• Simulate benchmark physics processes on different
  full detector designs.
• Software is easy to install, learn, use. 0 to
  analysis in 15 min.
• Goal is to allow software to be installed from CD
  or web with no external dependencies.
• Support via web based forums, tutorials, meetings.

5
LC Detector Full Simulation
Items highlighted in yellow represent common
standards
MC Event (stdhep)
Raw Event (lcio)
slic
Geometry (lcdd)
GEANT4
lcgo
6
Detector Variants

• Runtime XML format allows variations in detector
  geometries to be easily set up and studied
• Stainless Steel vs. Tungsten HCal sampling
  material
• RPC vs. GEM vs. Scintillator readout
• Layering (radii, number, composition)
• Readout segmentation (size, projective vs.
  nonprojective)
• Tracking detector technologies topologies
• TPC, Pixels, Silicon microstrip, SIT, SET
• Wedding Cake Nested Tracker vs. Barrel Cap
• Field strength
• Far forward MDI variants (0, 2, 14, 20 mr )

7
Detector Comparisons

• Each region has developed its own framework for
  detector simulations to reflect local tastes.
• ECFA ACFA detector simulations more tightly
  coupled to LDC and GLD detectors.
• Recognized early on the importance of closer
  collaboration to allow direct comparisons.

ECFA-ILC Mokka / Marlin C
ACFA-ILC Jupiter / Satellites Root
ALCPG slic / org.lcsim Java
8
LCIO

• Internationally agreed-upon event data model and
  persistency format.
• Defines interfaces for simulation and
  reconstruction objects.
• Implementations in Java, C Fortran.
• Allows mix-and-match at level of persistency or
  cross-language programming, and provides bindings
  across boundaries of event generation, detector
  response simulation, event reconstruction and
  analysis.

9
Motivation
Java, C, Fortran
Java, C, Fortran Geant3, Geant4
Java, C, Fortran
geometry (lcgo)
10
(No Transcript)
11
Example of Test Beam Analysis
12
Event Samples

• Have generated canonical data samples and have
  processed them through full detector simulations.
• simple single particles ?, ?, e, ?/- , n,
• composite single particles ?0,?, K0S ,?, ?, Z,
• Z Pole events comparison to SLD/LEP
• WW, ZZ, tt, qq, tau pairs, mu pairs, Z?, Zh
• beam pairs, muons, ??? hadrons, etc. backgrounds
• inclusive 1 ab-1 Standard Model sample
• Web accessible http//www.lcsim.org/datasets/ftp.h
  tml
• Additive at the detector hit level, with time
  offsets.
• Investigate effects of full backgrounds.

13
Reconstruction Toolkit

• Core reconstruction algorithms (track finding,
  fitting, calorimeter clustering, etc.) are in
  place.
• Interfaces defined for tasks, with many different
  plug--play implementations (e.g. calorimeter
  clustering).
• Standardized algorithm comparison tools.
• Analyses targeted to ReconstructedParticles
• Decouples interdependencies of different tasks.
• Allows comparisons between different algorithms
  or implementations.
• Easily swap in MC cheater to study effects of
  particular analysis task, independent of other
  tasks.
• Physics analyses can be developed and tested
  using fast Monte Carlo smearing, seamlessly
  transition to full reco.

14
Individual Particle Reconstruction

• Reconstruction is complex you have to get many
  individual steps right
• track finding, fitting extrapolation
• track-cluster matching
• MIP identification
• photon identification
• hadronic shower clustering (digital vs analog,
  scintillator vs gaseous readout)
• handling of displaced secondaries
• calibration of photons
• calibration of neutral hadrons
• E/p cut (including calibration)
• Algorithm development isn't about finding the
  "magic bullet" perfect algorithm. It's about
  iteratively
• finding the worst problems that are limiting
  performance
• fixing them
• hopefully seeing things improve a little
• finding the next worst problems

M. Charles
15
Individual Particle Reconstruction

• Algorithms being developed with minimal coupling
  to specific detector designs.
• Will allow full phase space of detector designs
  to be studied in a common framework.
• Finishing development of common infrastructure
  tools
• Calibration method for detector models
• Perfect PFA prescription
• Released reconstruction template
• Enables e.g. Cluster algorithm substitution, CAL
  hit/cluster accounting
• Migrating individual analyses into this framework
• Optimization Standardization of reconstructors
• Photon muon finders fairly mature, close to
  release (meeting _at_ Clermont-Farrand in July)
• Use of LCIO allows (in principle) exchange of
  algorithms across language, OS and regional
  boundaries.
• Analysis emphasis on dijet invariant mass
  resolution in physics events
• Currently ee- ??ZZ ?? (?n) (qq)
  (No jet combinatorics, uds) (2)
• Next in line ee- ??ZZ ?? (qq) (qq)
  ee- ??ZZ?n, WW?n (4)
• ??tt
  
  (6)
• ??tth
  
  (8)
• Plan to release canned physics analyses to
  reduce systematic uncertainties in e.g.
  jet-finding, combinatorics, constrained fits,

16
Performance Comparisons

• Reconstructing boson decays into dijets requires
  exceptional jet energy and direction resolution,
  is one of the driving forces in the current round
  of software development and will be used as a
  metric in the design of the detectors.
• Community benefits from close cooperation
  collaboration.
• Common input data samples (events in stdhep
  format)
• Common simulation/reconstruction output formats
• List of ReconstructedParticles in LCIO format
• WWS OC committee has proposed charging the
  Software Working Group to create an ILC Jet
  Energy Working Group.

see talk at LCWS07 for details.
17
The Grid

• Existing resources have proven sufficient to-date
  for event generation, detector response
  simulation and reconstruction/analysis (but is
  changing).
• Grid tools seem to be getting to the point where
  they are useful, so are beginning transition.
• Tools have been developed from the beginning to
  be grid friendly, i.e. static binaries, no db
  connections,
• Have developed (SBIR w/ Tech-X) Interactive
  Dataset Analysis on the Grid tools (as opposed to
  normal batch processing).
• Plug-in allows grid analysis from within JAS.

18
Resources for getting started

• http//lcsim.org/ Web Site
• Tutorials
• Software installation
• Using tools
• Simple Analysis Examples
• Developers Guide
• Datasets
• Documentation
• Confluence Wiki
• More tutorials
• More documentation
• Frequently asked Questions
• Users are encouraged to comment on, add to, or
  correct existing documentation
• https//jira.slac.stanford.edu/signup

19
Resources for getting started

• Discussion Forums
• http//forum.linearcollider.org/
• SLIC, org.lcsim
• Not recommended
• Spray E-mail to developers
• Banging head against wall
• Uninstall and reinstall software 3 times
• Recommended
• Post questions on the forum
• You will get faster answers
• You will get more accurate answers
• Others will benefit from seeing answers to your
  questions
• Discuss what you would like to do
• get feedback on best practices

20
Luminosity, Energy, Polarization

• 3.4 Extraction Line Energy Spectrometer
• 3.5 BPM-Based Energy Spectrometer
• 3.6 Polarimetry
• 3.7 Compton polarimeter backgrounds
• 3.8 Incoherent and coherent beamstrahlung
• 3.9 BeamCal and GamCal

See talk by M. Hildreth for details.
21
Vertex

• 4.1 Pixel Vertex Detector
• 4.2 Monolithic Pixel Detector Module
• 4.4 Vertex Detector Mech. Structures
• 4.5 Pixel-level Sampling CMOS VxDet

See talk by R. Lipton for details.
22
Tracking

• 5.2 GEM-based Forward Tracking
• 5.7 MPGD Readout for a TPC
• 5.8 Tracker Simulation and Alignment System
• 5.10 Long Shaping-Time Silicon Strip
• 5.13 Reconstruction Studies for the SiD Tracker
• 5.15 Calor based Tracking-Long-lived Particles
• 5.17 Thin silicon sensors
• 5.19 TPC signal digitization
• 5.21 2-D Readout of Silicon Strip Detectors

See talks by D. Peterson and T. Nelson for
details.
23
Calorimetry

• 6.1 Scintillator based Hadron Calorimeter
• 6.2 Scintillator Had Cal with SiPDs
• 6.4 Particle Flow Algorithm Development
• 6.5 Silicon-tungsten EM calorimeter
• 6.6 Digital Hadron Calorimetry w/ GEMs
• 6.9 Particle-Flow Algorithms and Simulations
• 6.10 ECAL Concepts for Particle Flow
• 6.14 Had Cal with Digital Readout (RPCs)
• 6.18 4th Concept Detector
• 6.19 Calorimeter and Muon ID
• 6.20 Scint/Cherenkov Rad Plates Cal w/ SiPMs

See talks by L. Xia , A. White , J. Hauptman R.
Frey for details.
24
Muon

• 7.2 Scintillator Based Muon System
• 7.5 Geiger-Mode APDs for Muon Systems
• 7.8 RPC and Muon System Studies

See talk by P. Karchin for details.
25
Other National Lab contributions

• SLAC provides infrastructure
• Geant4 full simulation (slic)
• Java Analysis Studio (JAS)
• Integrated Development Environment
• Analysis framework
• Wired event display
• LCIO in close collaboration with DESY LLR-Ecole
  Polytechnique.
• org.lcsim infrastructure
• MC event generation, detector simulation, data
  host.
• T. Johnson, R. Cassell, J. McCormick, N. Graf
• Contributions from T. Barklow, T. Maruyama, T.
  Nelson
• FNAL
• Muon reconstruction software (C. Milstene)
• Starting to study issues of forward tracking
  (tiling, segmentation, pattern recognition, etc.)
  as well as improvements to the tracking data
  model. (R. Kutschke, H. Wenzel, students)
• Possible GRID support.

26
Summary

• ALCPG Sim/Reco team supports an ambitious physics
  and detector simulation, reconstruction
  analysis effort.
• Goal is flexibility and interoperability, not
  technology or concept limited.
• Provides full data samples for ILC physics
  studies.
• Stdhep and LCIO files available on the web.
• Provides a complete and flexible detector
  simulation package capable of simulating
  arbitrarily complex detectors with runtime
  detector description.
• Reconstruction analysis framework exists, core
  functionality available, individual particle
  reconstruction template developed, various
  analysis algorithms implemented.
• Effort sorely lacking in manpower. Any additional
  support provides immediate detector performance
  results.

27
Additional Information

• lcsim.org - http//www.lcsim.org
• ILC Forum - http//forum.linearcollider.org
• Wiki - http//confluence.slac.stanford.edu/display
  /ilc/Home
• org.lcsim - http//www.lcsim.org/software/lcsim
• Software Index - http//www.lcsim.org/software
• Detectors - http//www.lcsim.org/detectors
• LCIO - http//lcio.desy.de
• SLIC - http//www.lcsim.org/software/slic
• LCDD - http//www.lcsim.org/software/lcdd
• JAS3 - http//jas.freehep.org/jas3
• AIDA - http//aida.freehep.org
• WIRED - http//wired.freehep.org

28

• Questions?

29
CALICE TCMT software work _at_ NIU

• Online monitoring package Monitor TCMT
  development, maintenance and support for data
  quality at TB site
• TCMT geometry drivers in Mokkamaintenance and
  support for TCMT simulation
• Official TCMT reconstruction packages in
  Marlinhit reconstruction and selection, in
  collaboration with S. Schmidt, B.Lutz et.al.
  (DESY)
• Digitization framework and modifiersnoise,
  x-talk, E-threshold hit ganging (simulation
  5x5cm2 cells --gt real 5x100cm2 strips)
• Support Java-based processing of CALICE data
  JAS3org.lcsimWiredDirected tree clustering in
  TB data, analysis and interactive event display
  (under way)

30
Purpose of DigiSim

• Goal a program to parametrically simulate the
  signal propagation and digitization processes for
  the ILC detector simulation? an important tool
  for comparing different detector technologies
• DigiSim role is to convert the simulated data
  (energy depositions and hit timings) into the
  same format AND as close as possible to real data
  from readout channels, while preserving all MC
  information from input data files
• As close as possible means that all known effects
  from digitization process should be taken into
  account, if possible cell ganging,
  inefficiencies, noise, crosstalks, hot and dead
  channels, non-linearities, attenuation, etc.
• Same reconstruction analysis software can be
  used for MC and real data
• DigiSim produces RawCalorimeterHits and/or
  calibrated (Sim)CalorimeterHits from (ideal)
  SimCalorimeterHits generated by Geant4.

31
DigiSim Status

• A digitization simulation package, DigiSim, has
  been developed at NICADD / NIU
• Java version released is full featured. Same
  configuration file as C (Marlin steering file)
• C version partly available. Same basic
  structure, but some functionality is missing
  (crosstalks and noise modeling), hopefully to be
  implemented soon
• Calibrated (Sim)CalorimeterHits can be directly
  compared to calibrated real data
• Both C and Java versions are available through
  official CVS serversC in the Calice CVS
  repository and Java in the LCSim CVS
  repository
• DigiSim can be run in either a stand-alone mode
  to produce persistent LCIO output, or as an
  on-the-fly preprocessor to reconstruction/analysis
  
• Documentation available from http//nicadd.niu.edu
  /digisim, including build instructions
• Comments are welcome lima_at_nicadd.niu.edu

32
(No Transcript)
33
VTX Simulation, Reconstruction, Analysis in
MARLIN Framework at LBNL
Physics Benchmarks
Jet Flavour Tagging CDF Vtx Fit HERAB Vtx Fit
Track Fit and Extrapolation
G4/Mokka
Sensor Simulation
lcio
Marlin
lcio
VTX Pattern Recognition
Cluster Reconstruction
PixelSim
lcio
ALS FNAL Beam Test Data
Sensor and Ladder Characterization
PixelAna
34
VTX Simulation Validation
LBNL ALS 1.5 GeV e- beam Run 056 Event 001
Cluster Pulse Height 1.5 GeV e- beam
Pattern Recognition and Track Fit Validation
Event Display of MokkaMarlin Simulation of VXD02
Change Cluster Size vs. Incidence Angle 1.5 GeV
e- beam
35
Development of Physics Analysis Tools

• Development of physics analysis tools
• Jet Flavour Tagging
• Vtx charge for 2f processes at
• highest energy, recovery
• Correction for b-jet with s.l.
• decay of heavy quark
• Topological Vtx Fit

Example in HAgbbbb at 1 TeV
Example Port of PUFITC developed for DELPHI at
LEP2 to MarlinReco framework using a C
wrapper
36
(No Transcript)
37
DHCAL slice 1m3 Prototype DAQ

• ANL,Boston University,FNAL,Imperial College
• Currently
• HAL based polling C-code for testing-debugging
  hardware.
• HAL based polling C-code for logging cosmic ray
  events at 2-4/min from 4 counter telescope.
• HAL based Interactive/menu based C-code for
  configuring front end ASIC via vme slow controls
  protocol.
• Routinely collecting cosmic ray data and
  producing ntuples (timestamp, X,Y,Z) for offline
  analysis with stack of 3-5 chambers.
• Root, Java, lcio offline monitoring and analysis
• Move to CALICE based framework with support for
  LCIO data formats.
• Data Archiving for offline analysis
• Run metadata and Hit pad-address/Timestamps
  stored on disk in binary format and converted to
  ntuples.
• Configuration data stored in SQL database.
• Event disk data will be converted to LCIO format.

38
Tracker Simulation and Alignment System

• U. Michigan group working on
• Alignment distortion simulations
• translation
• rotation
• twist
• expansion
• Evaluating resulting resolution degradation.
• Developing correction algorithms using frequency
  scanned interferometry measurements.

39
(No Transcript)
40
ILC RD at LA Tech

• Primarily concentrating on
• the Endcap Tracking Detector (ETD) in LDC
• Called FCH in the TELSA TDR
• Forward Tracking Studies
• Fast simulations of forward resolution using SGV
• Results shown at Snowmass, Vienna
• Detailed simulations of ETD in Mokka simulation
  framework
• Tracking in the forward region in Marlin
• Developing LDC geometry file for SLIC simulation
  framework
• Direct Mokka/SLIC comparisons
• Studying resolution requirements in the
  intermediate to forward angles.
• Studying effect of the TPC endplate on tracking
  resolution at intermediate angles

41
LA Tech on LDC

• Took part in drafting current Detector Outline
  Document (DoD)
• Co-Editor of Supplementary Tracking chapter
• Includes some simulations results obtained at LA
  Tech
• Simulation Wars Two ways of generating detector
  simulations
• SLAC STDHEP input gt SLIC GEANT interface gt
  org.lcsim reconstruction
• DESY STDHEP input gt MOKKA GEANT interface gt
  MARLIN reconstruction
• We have worked on geometry in both MOKKA and SLIC
  frameworks
• Both branches use LCIO file format for output
• E.g. Should be able to reconstruct SLIC output
  with MARLIN
• We are testing this at LA Tech.
• Detector RD collaborations
• Recently joined the LC-TPC collaboration
• Common interest in gaseous detectors (GEMs,
  micro-megas)
• Development of ETD cannot be independent of TPC
  endplate design
• New Forward Tracking Consortium formed with
  Indiana, Oklahoma (see later slide)

42
Recent LCRD Proposal

• Joint proposal with
• Oklahoma (Strauss),
• Indiana (van Kooten) and
• LA Tech (Sawyer, Greenwood, Wobisch Wells)
• First step in a possible Forward Tracking RD
  collaboration a la CALICE or LC-TPC.
• Funded at 31k (for all 3 institutions) for
  upcoming year.
• Continuation of previously described work at LA
  Tech
• Assistance from OK and IU in test beam,
  electronics development
• Year 3 of 3-year renewal cycle.
• Strong new effort from OK in forward tracking
  algorithms.
• Collaboration in detailed forward studies, incl.
  low angle forward tracking (i.e. FTD).
• Possible overlap with larger Southwest regional
  group developing upgrade RD plans for SLHC
• First planning meeting held in January, 2007
• Presentations from both LHC and ILC groups.

43
Simulations
LDC model for SLIC WIRED4 wireframe
representations
Muon reconstructed momentum in the endplate
region MOKKA vs SLIC
44
Testbeam Monitoring and Analysis Tools
(G.Mavromanolakis, Cambridge Univ./Fermilab)
Real time display of data quality and detector
status high performance tool for the shift
crew, essential during detector commissioning,
beam tuning and data taking More than 100 plots
3D event displays printouts/reports
available at the control room
CALICE testbeam at CERN
45
Testbeam Monitoring and Analysis Tools
(G.Mavromanolakis)
ECAL(W/scint.strip) testbeam at DESY
Testbeam runs continue at CERN in summer07 From
fall07 CALICE program moves to MTBF-FNAL
46
BeamCal GamCal

• Primarily University of Colorado BNL
• Tuning DID optimizing far-forward calorimetry
• Using slic/org.lcsim
• Close collaboration with FCAL collaboration.

47
Reconstruction Studies for the SiD Tracker

• Using slic org.lcsim to study track finding
  efficiency in the outer silicon tracker in SiD.
• Ported BaBar C Kalman fitter to Java org.lcsim
  framework for barrel geometries.

48
TPC digitization

• Converting Monte Carlo SimTrackerHits into
  TrackerHits corresponding to realistic readout.
• Targets LCIO, released as C Marlin processor.

49
Scintillator Had Cal with SiPDs

• Working within slic org.lcsim framework
• Studying overlapping scintillator tiles as ECal
• Developing ?2 techniques for ? ?0 ID.

50
Muon System Studies

• Working within slic and org.lcsim framework.
• Developed primarily by C. Milstene (FNAL).
• Pattern recognition within muon system for high
  pT muons, and within calorimeters for low pT.
• Kalman filter includes effects of multiple
  scattering, the integrated vxB term from the
  magnetic field and dE/dx.
• Stepper applies a realistic propagation at each
  step and allows hits to be collected in a
  narrower kinematic band.
• Muon Finder package released as org.lcsim Driver.

51
4th Concept Calorimetry

• Concept-specific simulation framework, not
  integrated into ECFA-ILC, ACFA-ILC or ALCPG
  environments.
• Full simulation is in place for HCAL and ECAL.
• Hits using Fluka (for calorimeter studies)
• Cerenkov and Scintillation photon production and
  propagation in the fibers fully simulated
• Full SDigits Digits Pattern Recognition chain
  implemented.

52
(No Transcript)
53
Conclusions

• Performance of Hadron Calorimeter is good
  (including pattern recognition) and INDEPENDENT
  OF ENERGY
• sE/E 34/ÖE (single hadrons)
• sE/E 40/ÖE (jets total energy no
  corrections)
• Expected to improve with jet finding (recovery of
  soft tracks and lost muons)
• Detector optimization not yet started (Cu vs Pb,
  fibers fraction)
• New Dual Readout EMCAL implemented
• sE/E 6/ÖE ECAL (very preliminary)
• sE/E 9/ÖE ECALHCAL (very preliminary)
• po-gtgg separation from t decay is under study
• Very high efficiency of muon finding in the midst
  of 4 jets (Muon Spectrometer)

54
Calor-based Tracking

• Using slic/org.lcsim
• Uses imaging EMCal to seed track-finding

55
Ongoing package updates performance improvements

• MIP stub finder
• tried several (lightly customized) clusterers
• dedicated MIP stub finder is in the works
• MIP stub handling in a fit
• Interoperability with other packages
• fitters, propagators, etc.
• mainly a question of infrastructure
• Handling of low Pt tracks
• again, improvements to infrastructure are
  critical
• Geometry optimization

56
Ongoing package updates new features

• Interface to proper digitization packages
• pixels (Nick Sinev)
• strips (Tim Nelson)
• Realistic and flexible virtual segmentation
• tile rings with wedges, hexagons, etc.
• Interface to the new geometry system
• need to better handle detectors with planar,
  overlapping silicon
• completely decouple from any particular geometry
• PFA and tracking tools
• track-cluster association
• look at tracks with fewer than 3 hits
• fake rate reduction tool

57
Experimental version based on new infrastructure
Charge deposition
Keeps true pulse height (collected charge)
events can be overlaid at this stage
Readout simulation
ADC output beam test data processing path
merges at this stage
Calibration
Ring
Cylinder
Wedge

Input to pattern recognition
Sensor
SensorType
Sensor
Sensor
Knows how to convert collection of
DigiTrackerHits into a measurement
Transformation3D
TrackerHitMaker
TrackFinder
TrackPoint
TrackNode
TrackNode
TrackNode
MipStub
TrackAnchor
Fitter
TrackerHit