Extracting overlapping gravitational-wave signals of galactic compact binaries: a mini review

TL;DR

This review discusses current methods for extracting overlapping gravitational-wave signals from galactic compact binaries in space-based detector data, highlighting iterative, global, hybrid, and machine learning approaches, and addressing the stochastic foreground.

Contribution

It provides a comprehensive overview of existing solutions for separating overlapping GW signals, aiming to inspire further research and improvements in this area.

Findings

Current solutions include iterative subtraction and global fitting methods.

Hybrid and machine learning techniques are increasingly used.

The stochastic foreground helps distinguish galactic signals from extragalactic backgrounds.

Abstract

Gravitational wave (GW) observations have provided a novel tool to explore the universe. In the near future, space-borne detectors will further open the window of low-frequency GW band where abundant sources exist and invaluable information for astrophysics, cosmology, and fundamental physics can be revealed. However, there are various new challenges in data analyses for space-borne detectors coming with the abundance of GW signals. For example, there are Galactic compact binaries (GCBs) with an overwhelming number that can produce continuous GW signals existing the entire mission time of detectors. The enormous overlapping GCB signals tangle and correlate with each other, and blend with other types of sources together in the observed data. Extracting source information from overlapping signals is one of the key problems for data analyses of space-borne detectors. In the paper, we present a review of currently available solutions for extracting overlapping GCB signals as thoroughly as possible aiming at promoting more interest in this question and inspiring further improvements. Current solutions can be roughly categorized by two classes, iterative subtraction and global fitting. There are diverse implementations of both strategies with enhancements focusing on different aspects. Meanwhile, the hybrid approach and the machine learning technique are also used in recent years. In the last, we also present an introduction of the stochastic foreground formed by unresolvable faint GCBs about its separation from extra-galactic backgrounds and its utility in exploring the properties of the Galaxy.