Structure of Reconstruction Program - PowerPoint PPT Presentation

About This Presentation
Title:

Structure of Reconstruction Program

Description:

Results from decoding of RAW data. after applying channel maps, calibration, drift-time relation... Pattern. Recognition. ESD. Old and New. Marco Cattaneo, ... – PowerPoint PPT presentation

Number of Views:36
Avg rating:3.0/5.0
Slides: 10
Provided by: marcoca1
Category:

less

Transcript and Presenter's Notes

Title: Structure of Reconstruction Program


1
Structure of Reconstruction Program
  • Towards BRUNEL

2
Some definitions
  • Reconstruction program transforms RAW data into
    Event Summary Data (ESD)
  • RAW data
  • IS digitised detector data as produced by the
    DAQ system
  • ADC counts, TDC counts
  • Integer data
  • IS NOT SICB raw data
  • Output of GEANT 3 transport step - entrance and
    exit points in detectors, energy loss...
  • IS NOT SICB digitisings
  • Output of SICB local tracking and digitisation
    step - wire numbers, signed drift distance

3
More definitions
  • Coordinates
  • Results from decoding of RAW data
  • after applying channel maps, calibration,
    drift-time relation...
  • 2D or 3D points in space, pulse heights, energy
    per cell
  • Floating point data
  • Determined locally to each subdetector
  • Intermediate data (usually not saved)
  • ESD
  • Results from pattern recognition and
    reconstruction
  • Tracks, Energy flow objects, PID
  • Determined globally
  • May require input from other subdetectors (e.g.
    tracks for RICH)
  • May combine results of several subdetectors (e.g.
    tracking, PID)
  • Implies sequencing of subdetectors, several passes

4
Old and New
SICB-Sim
Wrapped SICB
Entrance and exit points and energy loss in
detectors
Half way between raw data and coordinates. E.g.
wire number, signed drift distance
GEANT4 Gaudi simulation
Brunel
5
Towards BRUNEL
  • Input SICB raw data
  • Produce SICB digitisings in GAUDI
    FortranAlgorithm
  • i.e. wrapped FORTRAN. Can be several Algorithms,
    e.g. one per subdetector
  • Produce coordinates
  • Wrapped FORTRAN to SICB coordinates
  • GAUDI SicbConverters to Transient Event Data
    Model
  • Pattern recognition, reconstruction
  • Ideally all C
  • FortranAlgorithm possible if very modular

6
Structure
DigitisationA FortranAlgorithm
DigitisationB FortranAlgorithm
Coordinate ProducerA FortranAlgorithm
Reconstruction FortranAlgorithm
Coordinate ProducerB C Algorithm
Gaudi Transient Event Data
Reconstruction Algorithms
Sicb Converters
7
Caveat
  • In previous slide, SICB converters are
    unidirectional
  • FortranAlgorithms can produce data for later use
    in Fortran or C
  • C algorithms can only produce data for later
    use in C
  • Choice
  • Write also converters in opposite direction OR
  • Re-implement in C ALL reconstruction algorithms
    requiring input from upstream C algorithms

8
More Caveats
  • Ultimately, BRUNEL will require RAW data as input
  • From GEANT4 based simulation, from DAQ
  • Existing SICB digitisation will then be
    obsolete
  • No point to re-implement now in C for BRUNEL
  • Consider converters from SICB digitisings to
    transient raw data, followed by OO co-ordinate
    creation.
  • Existing wrapped co-ordinate creation code to be
    re-implemented with RAW data as input
  • No point to re-implement now in C with SICB
    digitisings as input

9
(No Transcript)
Write a Comment
User Comments (0)
About PowerShow.com