Overview
Coherent network analysis is addressing the problem of detection and reconstruction
of gravitational waves (GW) with networks of detectors.
It has been extensively
studied in the literature in application to detection of bursts
signals, which may be produced by numerous gravitational wave sources in the
Universe. In coherent methods, a statistic is built
up as a coherent sum over detector responses. In general, it is expected to be more
optimal (better sensitivity at the same false alarm rate) than the detection statistics
of the individual detectors that make up the network. Also coherent methods provide
estimators for the GW waveforms and the source coordinates in the sky.
The method we present (called coherent WaveBurst) is significantly different from
the traditional burst detection methods. Unlike coincident methods, which
first identify events in individual detectors by using an excess power statistic and than
require coincidence between detectors, the coherent WaveBurst method combines all
data streams into one coherent statistic constructed in the framework of the constrained
maximum likelihood analysis. Such an approach has significant advantages over the
coincident methods. First, the sensitivity of the method is not limited by the least
sensitive detector in the network. In the coherent WaveBurst method the detection is
based on the maximum likelihood ratio statistic which represents the total signal-to-noise
ratio of the GW signal detected in the network. Second, other coherent statistics, such as
the null stream and the network correlation coefficient can be constructed to distinguish
genuine GW signals from the environmental and instrumental artifacts. Finally, the
source coordinates of the GW waveforms can be reconstructed.
- Extracted from:
S. Klimenko, I. Yakushin , A. Mercer, G. Mitselmakher
, Coherent method for detection of gravitational
wave bursts, Class. Quantum Grav. 25, 114029 (2008): link to
CQG, link to
arXiv
Advanced LIGO detectors has started their operation
at unprecedented sensitivity targeting first detection
of gravitational waves from astrophysical sources.
A more robust detection of gravitational waves is
anticipated in the next few years as the advanced LIGO
reaches its designed sensitivity and the other advanced
detectors Virgo Kagra and LIGO-India come
online. Numerous GW signals, expected to be observed
by the advanced detectors (
40 binary neutron star and
possible black hole mergers per year will begin our
exploration of the gravitational-wave sky and start the
era of the gravitational wave astronomy.
The advanced detectors target detection of GW transients
for a wide range of promising astrophysical
sources including: various types of gamma-ray bursts,
core-collapse supernovae, soft-gamma repeaters, cosmic
strings, late inspiral and mergers of compact binaries,
ring-downs of perturbed neutron stars or black holes, and
as-yet-unknown systems. Most of these sources are
difficult to model, due to their complicated dynamics and
because the equation of state of matter at neutron star
densities is not known. Therefore, the search algorithms
have been developed for detection of GW transients,
or bursts of GW radiation in the detector bandwidth,
with no or little assumptions on the source models.
There are two different ways the GW searches are conducted:
in real time and searches on the archived data.
The objective of the real time burst search is the
identification and reconstruction of significant event
candidates with low latency (within few minutes).
The reconstructed sky location can be promptly shared with the
partner telescopes, which search for a coincident
electromagnetic (EM) counterpart A prominent source
for such join observation is a merger of compact binary
objects where one of the companions (or both) is a neutron star.
Such mergers may produce several EM signals:
gamma-ray busts (GRB), GRB afterglow, kilonova, etc,
which will fade away with the time scales ranging from
seconds to days A small fraction of such mergers
(when the GRB beam is pointing at us) can be independently
detected by the gamma-ray telescopes and associated
with a GW signal by the time of the event. However,
most of the compact binary mergers require a prompt
sky localization with the GW detectors and follow-up
EM searches for possible afterglow. Similar observations
can be performed for the galactic events such as supernovae
or soft-gamma-repeaters, which may produce both
the EM and neutrino counterparts. On contrary, the
objective of the archived burst analysis is to establish
a significance of observed events and identify their progenitors.
Such analysis requires detail background studies and
accurate reconstruction of the source parameters,
which may not be readily available with low latency.
Both types of searches and the sky localization studies
have been performed with the baseline burst algorithm
coherent WaveBurst (cWB) used in the analysis of
data form the initial instruments. In this paper
we describe the improvements of the cWB algorithm,
which is currently used both for the real time burst search
and several archived searches with the networks of
advanced detectors. This second generation cWB algorithm
includes several novelties. The time-frequency analysis
has been updated with a novel time frequency
transform which improved the waveform reconstruction.
It also significantly improved the computational performance
of the algorithm, enabling a robust low latency
operation. The data conditioning (whitening, removal of
the spectral artifacts, etc) has been enhanced with the
data regression algorithms Fast reconstruction of
the chirp mass has been introduced to enable rapid
identification of the compact binary coalescence (CBC)
sources. The extensive sky localization studies have been
performed
- Extracted from:
S. Klimenko, G. Vedovato, M. Drago, F. Salemi, V. Tiwari, G.A. Prodi,
C. Lazzaro, K. Ackley, S. Tiwari, C.F. Da Silva, and G. Mitselmakher,
Method for detection and reconstruction of gravitational wave
transients with networks of advanced detectors, Phys. Rev. D 93,
042004 (2016): link to
PRD,
link to arXiv
References