Hierarchical searches for subsolar-mass binaries and the third-generation gravitational wave detector era

TL;DR

This paper introduces a hierarchical search method for detecting subsolar-mass binaries in third-generation gravitational wave data, significantly reducing computational costs while improving detection sensitivity.

Contribution

It presents a novel two-stage hierarchical search strategy optimized for long-duration signals from ultra-compact objects in the era of third-generation detectors.

Findings

Lowering the search frequency to 35 Hz increases SNR by 6%.

Detection volume improves by 10-20%.

Computational costs are reduced by a factor of 2.5.

Abstract

The detection of gravitational waves (GWs) from coalescing compact binaries has become routine with ground-based detectors like LIGO and Virgo. However, beyond standard sources such as binary black holes and neutron stars and neutron star black holes, no exotic sources revealing new physics have been discovered. Detecting ultra-compact objects, such as subsolar mass (SSM) compact objects, offers a promising opportunity to explore diverse astrophysical populations. However, searching for these objects using standard matched-filtering techniques is computationally intensive due to the dense parameter space involved. This increasing computational demand not only challenges current search methodologies but also poses significant obstacles for third-generation (3G) ground-based GW detectors. In the 3G era, signals may last tens of minutes, and detection rates could reach one per minute, requiring efficient search strategies to manage the computational load of long-duration signals. In this paper, we demonstrate a hierarchical search strategy designed to address the challenges of searching for long-duration signals, such as those from SSM compact binaries, and the anticipated issues with 3G detectors. We show that by adopting optimization techniques in a two-stage hierarchical approach, we can efficiently search for the SSM compact object in the current LIGO detectors. Our preliminary results show that conducting matched filtering at a lower frequency of 35 Hz improves the signal-to-noise ratio by 6% and enhances the detection volume by 10-20%, compared to the standard two-detector PyCBC search. This improvement is achieved while reducing computational costs by a factor of 2.5.