1 / 22

The QPipeline heirarchical coherent search for gravitational wave bursts

The QPipeline heirarchical coherent search for gravitational wave bursts. Shourov K. Chatterji INFN Sezione di Roma / Caltech LIGO Laboratory for the LIGO Scientific and Virgo Collaborations 7 th Eduardo Amaldi Conference on Gravitational Waves 2007 July 8-14 Sydney, Australia. Abstract.

clairea
Télécharger la présentation

The QPipeline heirarchical coherent search for gravitational wave bursts

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. The QPipeline heirarchical coherent search for gravitational wave bursts Shourov K. Chatterji INFN Sezione di Roma / Caltech LIGO Laboratory for the LIGO Scientific and Virgo Collaborations 7th Eduardo Amaldi Conference on Gravitational Waves 2007 July 8-14 Sydney, Australia

  2. Abstract LIGO/Virgo Scientific Collaborations meeting - 2007 March 19

  3. Note This work will be presented as a poster. It will also need to be significantly condensed. This represents the material from which it will be taken.

  4. S5 QPipeline search • Search the first year of S5 data (through 2006 Nov 14 18:00 UTC) • Use H1 and H2 to improve the sensitivity to gravitational-wave bursts and the rejection of instrumental artifacts • There are two primary components of this analysis • H1H2 double coincident search for combined excess signal energy followed by H1H2 null stream consistency test • H1H2L1 triple coincident search for time frequency coincidence between H1H2 triggers and L1 triggers • Upper limits will be determined using the loudest event statistic • The most significant ~100 events will be followed up • QScan: Scan auxiliary detector and environmental channels for statistically significant signal content in coincidence with gravitational-wave signal • QEvent and XPipeline: Coherent tests for consistency with a direction on the sky if data is available from a sufficient number of detectors • For more information see the project summary page and technical documentation LIGO/Virgo Scientific Collaborations meeting - 2007 March 19

  5. Single detector QPipeline • Multiresolution time-frequency search for statistically significant excess signal energy • Projects whitened data onto an overlapping basis of sinusoidal Gaussians characterized by central time, central frequency, and Q (ratio of central frequency to bandwidth) • The template bank is constructed using a maximum mismatch approach similar to the matched filtering approach • The search is equivalent to a matched filter search for waveforms that are sinusoidal Gaussians after whitening • The reported normalized energy Z is a measure of event significance and is simply twice the squared SNR r that would be reported by a matched filter search LIGO/Virgo Scientific Collaborations meeting - 2007 March 19

  6. Collocated detectors • Special easy to understand case of coherent methods • Applicable to the two LIGO Hanford detectors • Subset of the general case • Produce hybrid ‘H’ detector than makes optimal use of H1 and H2 • Exhibits many of the same features as the general case • Optimal combination to maximize detectability • Null combination to test for consistency • But there are some differences • Coherent combinations are independent of source direction • Computationally much cheaper than general case • Cannot fully recover source information • Forms the first stage of a hierarchical coherent search • Must be careful of correlated events due to common environment LIGO/Virgo Scientific Collaborations meeting - 2007 March 19

  7. H1H2 QPipeline • The H1 and H2 can be combined to form two new detector data streams H+ The optimal linear combination that maximizes the signal to noise ratio of potential signals. • Frequency dependent weighting factors are inversely proportional to power spectral density S in each bin • Resulting SNR is the quadrature sum of SNRs H- The null stream, which should be consistent with noise in the case of a true gravitational-wave • Frequency independent weighting factors LIGO/Virgo Scientific Collaborations meeting - 2007 March 19

  8. H1H2 QPipeline • Calibration uncertainty can produce a significant residual null stream signal for strong gravitational waves • Compare null stream significance with the significance expected on the assumption of uncorrelated detectors • In practice, Z0 this is given by the normalized energy of H1+H2 • H- events are significant if • The parameter a determines the false veto rate • The parameter b accounts for calibration uncertainty • Initial tuning studies on glitches, hardware injections, and software injections (with simulated uncertainties) suggests a=10, b=0.05 • Veto significant H+ events that overlap in time and frequency with a significant H- event. LIGO/Virgo Scientific Collaborations meeting - 2007 March 19

  9. H1H2 example: Inspiral hardware injection at 5 Mpc H1 H2 H+ yields ~10 percentincrease in SNR H- H+ H- consistent withdetector noise LIGO/Virgo Scientific Collaborations meeting - 2007 March 19

  10. H1H2 example: time shifted glitch H1 H2 Coincident H1H2 glitch In time-shifted data set H- H+ Significant H- contentindicates inconstistency LIGO/Virgo Scientific Collaborations meeting - 2007 March 19

  11. H1H2 example: Inspiral at 0.1 Mpc H1 H2 Residual much smallerthan original signal H- H+ Residual signal due tocalibration uncertainty LIGO/Virgo Scientific Collaborations meeting - 2007 March 19

  12. Analyzable data from LIGO S5 run Analyzable good quality data from the first calendar year of LIGO’s fifth science run Previous LSC burst searchesonly focused on the H1H2L1triple coincident data set H1H2 246.1 65.7% The double coincidentH1H2 data set providesa significant increase induty cycle LIGO/Virgo Scientific Collaborations meeting - 2007 March 19

  13. Generalized coherent follow-up • A fully coherent follow-up to candidate events can be performed when data is available from three or more non-aligned detectors • Analogous to collocated H1H2 analysis • Produce linear combination of time-shifted detector data that maximizes the signal to noise ratio of potential signals • Produce linear combination(s) that contain no signal • Compare null streams with expected null stream based on the assumption of uncorrelated detector data • Produces consistency sky maps LIGO/Virgo Scientific Collaborations meeting - 2007 March 19

  14. Coherent follow-up Generalize to arbitrary number of detectors Coherent sum: Find linear combinations ofdetector data that maximizesignal to noise ratio data = response x signal + noise X = F h + n coherent null N-2 dimensionalnull space detector data Null sum: Linear combinations ofdetector data that cancelthe signal provide usefulconsistency tests. coherent sum 2 dimensionalsignal space PRD 74 (2006) 082005, gr-qc/0605002 LIGO/Virgo Scientific Collaborations meeting - 2007 March 19

  15. Note The following plots were generated with simulated LIGO/Virgo data. They are being regenerated with real LIGO/Virgo data from project 2b, which includes a secret time shift. They will be relabeled as “time shifted typical LIGO and Virgo data”.

  16. Example null stream consistency test c2 / DOF consistency with a GWB as a function of direction for a simulated supernova (~1 kpc) Interference fringes from combining signal in two detectors. • True source location: • intersection of fringes • c2 / DOF ~ 1 GWB: Dimmelmeier et al. A1B3G3 waveform, Astron. Astrophys. 393 523 (2002) , SNR = 20 Network: H1-L1-Virgo, design sensitivity LIGO/Virgo Scientific Collaborations meeting - 2007 March 19

  17. Example null stream consistency test • GWBs: • 3 core-collapse supernova waveforms. • Dimmelmeier, Font, & Müller, A&A 393 523-542 (2002). • Pick one DFM and add to each detector data stream. • Glitches: • Inject a different supernova waveform into each detector • Use same time delays, amplitudes as a GWB. Pathological glitches! • Detector Network: • LIGO-Virgo network @ design sensitivity LIGO/Virgo Scientific Collaborations meeting - 2007 March 19

  18. Example consistency sky maps Simulated gravitational-wave burst Simulated coincident glitch Phys.Rev. D74 (2006) 082005, gr-qc/0605002 LIGO/Virgo Scientific Collaborations meeting - 2007 March 19

  19. GWB and glitch populations clearly distinguished for SNR > 10-20. Similar to detection threshold in LIGO. Example: 5000 bursts vs. 5000 glitches • One point from each simulation. • sky position giving strongest cancellation rrms = 100 50 Incoherent Energy 20 10 5 Null = Incoherent + Correlated LIGO/Virgo Scientific Collaborations meeting - 2007 March 19

  20. Receiver operator characteristic LIGO/Virgo Scientific Collaborations meeting - 2007 March 19

  21. Hierarchical search approach • Coherent methods are powerful but computationally costly • The H1H2 coherent search is computationally inexpensive and can also be used as a first step in a hierarchical coherent search • Candidate events can be followed up using with coherent consistency tests that consider data from all available detectors • Coincident methods can also be operated at low significance threshold to generate a set double coincident candidate events for follow-up by coherent tools • Eliminates the difficulties of tuning the coincident methods • Achieves a sensitivity comparable to coherent methods with a cost comparable to coincident methods LIGO/Virgo Scientific Collaborations meeting - 2007 March 19

  22. Comments on coherent methods • Null stream veto limited by least sensitive detector, e.g.: • Successfully vetoes coincident H1H2 glitches • Successfully vetoes H2 only glitches • Successfully vetoes strong H1 only glitches thatwere strong enough to have been seen in H2 • Avoids false dismissal of gravitational waves thatare strong enough to been seen in H1 but not H2 • At best, the coherent sum provides a improvement • Only if all N detectors have equal sensitivity • Only if all N detectors see the same signal • The same statistic is not best for all three tasks • Due to the presence of glitches, coherent null streams may provide a much greater improvement in sensitivity and detection confidence than the coherent sum LIGO/Virgo Scientific Collaborations meeting - 2007 March 19

More Related