LSC-Virgo Burst Analysis Working Group

Navigation

Burst Group Home
Wiki
ExtTrig Home
CBC Group Home
LSC, LIGO, Virgo

Documents

'How-to' docs
Agendas/minutes
[2008, 2007, earlier]
Talks/Posters [pre-wiki]
Papers
Paper plans
White papers
ViewCVS

Investigations

Analysis projects
Old notebook [General, S2, S3, S4, S5]
Virgo workarea
External collabs

Review

Main review page
Telecons

Review Committee Meeting Monday November 17 2008 08:00 Pacific / 11:00 Eastern

Minutes: Monday November 17 2008 08:00 Pacific / 11:00 Eastern

Agenda and Contact Info

Agenda

  1. Announcements
  2. Update on status of S5 1st Year Burst Search Paper [ HTML ]
  3. Reports from Burst Review Teams
    • Report from the Block Normal review team [ PDF ]
  4. A.O.B.

Contact Info

InterCall telecon service:
Phone: 1-866-380-5536, participant code: 435 672 9587 #
International callers ++1-816-249-4731 with same code
International access numbers [ HTML ]
Handy participant command: Press *6 ["*M"] to mute, #6 to unmute

Minutes

  1. Announcements
    • Please schedule things that need to be done before LSC meeting.
  2. Update on status of S5 1st Year Burst Search Paper [ HTML ]
    • No report -- we're supposed to get the paper by Wednesday.
  3. Reports from Burst Review Teams
    • Report from the Block Normal review team [ PDF ]
      • Looked at entire pipeline, scripts, condor logs.
      • Data conditioning. Examined each stage. A few minor issues, but nothing outstanding that would invalidate result.
      • Slight fall-off on first Rate vs. TimeLag plot Not clear why this is, but it is only when very few cuts. (Fig. 7.)
      • Figure 8: should be two plots: one for LF second for HF.
      • Figure 9: shows bad frequency behavior for loud injections. Centroid in time being calculated incorrectly when clustering events within a detector. Error means events are lost in coincidence.
        Code change has been made. Is in CVS. Do we want to redo the data analysis? It won't affect overall S5 sensitivity because mixing all pipelines. Would take about 1 month per set to redo MDCs, a week to do on-source and background.
        Proposal: leave things as they are for computing the upper limit. But fix the code and rerun to make sure that no loud events were actually missed. "We notices that there was a problem with loud injections ... this was traced to ((explain problem))... the UL is still valid but we fixed the code and checked that there were no loud events that we missed."
      • Calibration.
      • Live-time. Spot-checks of Cat2 and Cat3 times. For Cat1 checked to make sure that the pipeline correctly accounts for all the time. For Cat2 and Cat3 vetoes: cause a loss of live-time rather than a loss of efficiency. Corrections need to be done in rate plots. If centroid of any trigger lied in the vetoed time, was rejected.
      • Efficiency: need to reject MDC events if in Cat2 or Cat3 time. Currently these flags are ignored for Cat2 or Cat3 time so efficiency goes to unity.
        Jolien's position: injections must be done in precisely the same way as the on-source triggers or else efficiency is suspect. Keith can do this because it is only post-processing. Question is what the burst group really wants. Need to figure out this issue in terms of how ETGs are combined.
      • Transition from BN to CP is done in pipeline. It has been checked. MDCs from reviewed code. Need to take h(t) to DARM_ERR. Calls same routines as the MDC code that was reviewed. Log files are read in for post-processing using the same code as was reviewed. (Log files not used until post-processing.)
      • Figure 8 has only background. There is a large CP outlier. It was looked at but no obvious reason for this event. Perhaps we should look at this event a little more.
      • Calibration errors: used mis-calibrated MDCs to see effect of the calibration errors at the 10% level. Both BN and CP. This gives final calibration error estimate.
      • TODO:
        1. Figures for Gamma plot, by frequency.
        2. How to do the livetime properly
        3. Corrected code and do zero-lag analysis.
        4. Provide discussion of background outlier.
  4. A.O.B.
$Id: minutes-2008-11-17.html,v 1.4 2008/11/17 17:05:56 jolien Exp $