-
Fast and faithful interpolation of numerical relativity surrogate waveforms using meshfree approximation
Authors:
Lalit Pathak,
Amit Reza,
Anand S. Sengupta
Abstract:
Several theoretical waveform models have been developed over the years to capture the gravitational wave emission from the dynamical evolution of compact binary systems of neutron stars and black holes. As ground-based detectors improve their sensitivity at low frequencies, the real-time computation of these waveforms can become computationally expensive, exacerbating the steep cost of rapidly rec…
▽ More
Several theoretical waveform models have been developed over the years to capture the gravitational wave emission from the dynamical evolution of compact binary systems of neutron stars and black holes. As ground-based detectors improve their sensitivity at low frequencies, the real-time computation of these waveforms can become computationally expensive, exacerbating the steep cost of rapidly reconstructing source parameters using Bayesian methods. This paper describes an efficient numerical algorithm for generating high-fidelity interpolated compact binary waveforms at an arbitrary point in the signal manifold by leveraging computational linear algebra techniques such as singular value decomposition and meshfree approximation. The results are presented for the time-domain \texttt{NRHybSur3dq8} inspiral-merger-ringdown (IMR) waveform model that is fine tuned to numerical relativity simulations and parameterized by the two component-masses and two aligned spins. For demonstration, we target a specific region of the intrinsic parameter space inspired by the previously inferred parameters of the \texttt{GW200311\_115853} event -- a binary black hole system whose merger was recorded by the network of advanced-LIGO and Virgo detectors during the third observation run. We show that the meshfree interpolated waveforms can be evaluated in $\sim 2.3$ ms, which is about $\times 38$ faster than its brute-force (frequency-domain tapered) implementation in the \textsc{PyCBC} software package at a median accuracy of $\sim \mathcal{O}(10^{-5})$. The algorithm is computationally efficient and scales favourably with an increasing number of dimensions of the parameter space. This technique may find use in rapid parameter estimation and source reconstruction studies.
△ Less
Submitted 28 March, 2024;
originally announced March 2024.
-
Pinpointing coalescing binary neutron star sources with the IGWN, including LIGO-Aundha
Authors:
Sachin R. Shukla,
Lalit Pathak,
Anand S. Sengupta
Abstract:
LIGO-Aundha (A1), the Indian gravitational wave detector, is expected to join the IGWN and begin operations in the early 2030s. We study the impact of A1 on the accuracy of determining the direction of incoming transient signals from coalescing BNS sources with moderately high SNRs. It is conceivable that A1's sensitivity, effective bandwidth, and duty cycle will improve incrementally through mult…
▽ More
LIGO-Aundha (A1), the Indian gravitational wave detector, is expected to join the IGWN and begin operations in the early 2030s. We study the impact of A1 on the accuracy of determining the direction of incoming transient signals from coalescing BNS sources with moderately high SNRs. It is conceivable that A1's sensitivity, effective bandwidth, and duty cycle will improve incrementally through multiple detector commissioning rounds to achieve the desired `LIGO-A+' design sensitivity. For this purpose, we examine A1 under two distinct noise PSDs. One mirrors the conditions during the O4 run of the LIGO Hanford and Livingston detectors, simulating an early commissioning stage, while the other represents the A+ design sensitivity. We consider various duty cycles of A1 at the sensitivities mentioned above for a comprehensive analysis. We show that even at the O4 sensitivity with a modest $20\%$ duty cycle, A1's addition to the IGWN leads to a $15\%$ reduction in median sky-localization errors ($ΔΩ_{90\%}$) to $5.6$~sq.~deg. At its design sensitivity and $80\%$ duty cycle, this error shrinks further to $2.4$~sq.~deg, with 84\% sources localized within a nominal error box of $10$~sq.~deg! Even in the worst-case scenario, where signals are sub-threshold in A1, we demonstrate its critical role in reducing the localization uncertainties of the BNS source. Our results are obtained from a large Bayesian PE study using simulated signals injected in a heterogeneous network of detectors using the recently developed meshfree approximation aided rapid Bayesian inference pipeline. We consider a seismic cut-off frequency of 10 Hz for all the detectors. We also present hypothetical improvements in sky localization for a few GWTC-like events injected in real data and demonstrate A1's role in resolving the degeneracy between the luminosity distance and inclination angle parameters.
△ Less
Submitted 16 April, 2024; v1 submitted 27 November, 2023;
originally announced November 2023.
-
Prompt sky localization of compact binary sources using a meshfree approximation
Authors:
Lalit Pathak,
Sanket Munishwar,
Amit Reza,
Anand S. Sengupta
Abstract:
The number of gravitational wave signals from the merger of compact binary systems detected in the network of advanced LIGO and Virgo detectors is expected to increase considerably in the upcoming science runs. Once a confident detection is made, it is crucial to reconstruct the source's properties rapidly, particularly the sky position and chirp mass, to follow up on these transient sources with…
▽ More
The number of gravitational wave signals from the merger of compact binary systems detected in the network of advanced LIGO and Virgo detectors is expected to increase considerably in the upcoming science runs. Once a confident detection is made, it is crucial to reconstruct the source's properties rapidly, particularly the sky position and chirp mass, to follow up on these transient sources with telescopes operating at different electromagnetic bands for multi-messenger astronomy. In this context, we present a rapid parameter estimation (PE) method aided by mesh-free approximations to accurately reconstruct properties of compact binary sources from data gathered by a network of gravitational wave detectors. This approach builds upon our previous algorithm [L. Pathak et al., Fast likelihood evaluation using meshfree approximations for reconstructing compact binary sources, https://journals.aps.org/prd/abstract/10.1103/PhysRevD.108.064055, Phys. Rev. D 108, 064055 (2023)] to expedite the evaluation of the likelihood function and extend it to enable coherent network PE in a ten-dimensional parameter space, including sky position and polarization angle. Additionally, we propose an optimized interpolation node placement strategy during the start-up stage to enhance the accuracy of the marginalized posterior distributions. With this updated method, we can estimate the properties of binary neutron star (BNS) sources in approximately 2.4~(2.7) min for the \TaylorF~(\texttt{IMRPhenomD}) signal model by utilizing 64 CPU cores on a shared memory architecture. Furthermore, our approach can be integrated into existing parameter estimation pipelines, providing a valuable tool for the broader scientific community. We also highlight some areas for improvements to this algorithm in the future, which includes overcoming the limitations due to narrow prior bounds.
△ Less
Submitted 17 February, 2024; v1 submitted 13 September, 2023;
originally announced September 2023.
-
Does the speed of gravitational waves depend on the source velocity?
Authors:
Rajes Ghosh,
Sreejith Nair,
Lalit Pathak,
Sudipta Sarkar,
Anand S. Sengupta
Abstract:
The second postulate of special relativity states that the speed of light in vacuum is independent of the emitter's motion. The test of this postulate so far remains unexplored for gravitational radiation. We analyze data from the LIGO-Virgo detectors to test this postulate within the ambit of a model where the speed of the emitted GWs ($c'$) from a binary depends on a characteristic velocity…
▽ More
The second postulate of special relativity states that the speed of light in vacuum is independent of the emitter's motion. The test of this postulate so far remains unexplored for gravitational radiation. We analyze data from the LIGO-Virgo detectors to test this postulate within the ambit of a model where the speed of the emitted GWs ($c'$) from a binary depends on a characteristic velocity $\tilde{v}$ proportional to that of the reduced one-body system as $c' = c + k\, \tilde{v}$, where $k$ is a constant. We have estimated the upper bound on the 90\% credible interval over $k$ to be ${k \leq 8.3 \times {10}^{-18}}$, which is several orders of magnitude more stringent compared to previous bounds obtained from electromagnetic observations. The Bayes' factor supports the second postulate with a strong evidence that the data is consistent with the null hypothesis $k = 0$, upholding the principle of relativity for gravitational interactions.
△ Less
Submitted 13 December, 2023; v1 submitted 28 April, 2023;
originally announced April 2023.
-
Fast likelihood evaluation using meshfree approximations for reconstructing compact binary sources
Authors:
Lalit Pathak,
Amit Reza,
Anand S. Sengupta
Abstract:
Several rapid parameter estimation methods have recently been advanced to deal with the computational challenges of the problem of Bayesian inference of the properties of compact binary sources detected in the upcoming science runs of the terrestrial network of gravitational wave detectors. Some of these methods are well-optimized to reconstruct gravitational wave signals in nearly real-time neces…
▽ More
Several rapid parameter estimation methods have recently been advanced to deal with the computational challenges of the problem of Bayesian inference of the properties of compact binary sources detected in the upcoming science runs of the terrestrial network of gravitational wave detectors. Some of these methods are well-optimized to reconstruct gravitational wave signals in nearly real-time necessary for multi-messenger astronomy. In this context, this work presents a new, computationally efficient algorithm for fast evaluation of the likelihood function using a combination of numerical linear algebra and mesh-free interpolation methods. The proposed method can rapidly evaluate the likelihood function at any arbitrary point of the sample space at a negligible loss of accuracy and is an alternative to the grid-based parameter estimation schemes. We obtain posterior samples over model parameters for a canonical binary neutron star system by interfacing our fast likelihood evaluation method with the nested sampling algorithm. The marginalized posterior distributions obtained from these samples are statistically identical to those obtained by brute force calculations. We find that such Bayesian posteriors can be determined within a few minutes of detecting such transient compact binary sources, thereby improving the chances of their prompt follow-up observations with telescopes at different wavelengths. It may be possible to apply the blueprint of the meshfree technique presented in this study to Bayesian inference problems in other domains.
△ Less
Submitted 1 December, 2023; v1 submitted 6 October, 2022;
originally announced October 2022.
