Linear Collider Forum

Home » Analysis and Reconstruction » EUDET Telescope » Important steps toward the EUDET telescope analysis
Important steps toward the EUDET telescope analysis [message #721] Mon, 19 February 2007 06:17
Messages: 66
Registered: January 2007
Location: INFN - Roma3

Dear All!
This short post to announce that we are very close to the first release of the EUTelescope Marlin processors. The code comes with doxygen documentation, but since we still don't have a place to publish it, I will try to describe here what has been already implemented and what is still missing. For the time being I tried to avoid ROOT linkage and I'm using AIDA for histogramming.

This is what we already have:

  • EUTelSucimaImagerReader. This is just a test to implement the on-line conversion of the standard SUCIMA DAQ format. Marlin reads an event in the SUCIMA DAQ format and converts it on the fly into a LCEvent. In principle we can implement a reader also for the IRES USB Imager board.
  • EUTelPedestalNoiseProcessor. This processor calculates the pedestal and the noise value of each pixel of each detector in the telescope. Few algorithm have been implemented and waiting to be fully debugged. The results of this processor are saved into a "Condition" file (it can also be a database entry, but not sure it is really needed).
  • EUTelCalibrateEventProcessor. This processor removes the pedestal from the current rawdata and eventually suppress the common mode. The output of this processor is a collection of TrackerData object, one per detector, with the data ready for the cluster search.
  • EUTelClusteringProcessor. This processor looks into each detector TrackerData for group of pixels passing the cluster selection cuts. The output of this processor is a collection of TrackerData containing only the clusters.

What is missing:

  • EUTelAutoPedestalProcessor. This processor will be used with self-biased pixel detectors in which there is actually no need for a real pedestal calculation. The pedestal and noise values for each pixel can be initialize to a reasonable value and then, during the first N events updates to the correct value.
  • EUTelClusterMergingProcessor. This processor will calculate the distance between clusters belonging to the same detector. In the case the two clusters are touching each other, something should be done to separate them. Probably not really needed in a telescope setup where the number of clusters per plain should be sufficiently low.
  • EUTelDataQualityMonitor. This processor should take as input the clusters provided by the EUTelClusteringProcessor and fill in sets of histograms/profiles/distributions... to cross-check the data quality at the detector level. Typical examples are the SNR for each detector, the beam profile, the eta function...
  • All the remaining track finding / fitting stuff.
  • Interfacing to the geometry database.

You can download the source code from here or via anonymous CVS. Once you have the source code, uncompress it into $(MARLIN)/packages and type make from the Marlin top folder.

Type make doc to produce the documentation.

To produce some example files browse to the $(MARLIN)/packages/Eutelescope/test and read the README files.

Please give it a try and let me known what you think about. Your feedback is very well appreciated!

Best regards,


Read Message
Previous Topic:RunHeader EUDET specific parameters
Next Topic:A Marlin processor for pedestal calculation
Goto Forum:

Current Time: Sun Sep 15 05:22:17 Pacific Daylight Time 2019
.:: Contact :: Home ::.

Powered by: FUDforum 3.0.1.
Copyright ©2001-2010 FUDforum Bulletin Board Software