学术文献库

Extracting gravitational wave (GW) signals from individual Galactic binaries (GBs) against their self-generated confusion noise is a key data analysis challenge for space-borne detectors operating in the $\approx 0.1$ mHz to $\approx 10$ mHz range. Given the likely prospect that there will be multiple such detectors, namely LISA, Taiji, and Tianqin, with overlapping operational periods in the next decade, it is important to examine the extent to which the joint analysis of their data can benefit GB resolution and parameter estimation. To investigate this, we use realistic simulated LISA and Taiji data containing the set of $30\times 10^6$ GBs used in the first LISA data challenge (Radler), and an iterative source extraction method called GBSIEVER introduced in an earlier work. We find that a coherent network analysis of LISA-Taiji data boosts the number of confirmed sources by $\approx 75\%$ over that from a single detector. The residual after subtracting out the reported sources from the data of any one of the detectors is much closer to the confusion noise expected from an ideal, but infeasible, multisource resolution method that perfectly removes all sources above a given signal-to-noise ratio threshold. While parameter estimation for sources common to both the single detector and network improves broadly in line with the enhanced signal to noise ratio of GW sources in the latter, deviation from the scaling of error variance predicted by Fisher information analysis is observed for a subset of the parameters.