Data Analysis Methods for the Ground-based and Space-borne Gravitational Wave Detector Network

It has been exactly ten years since humans first directly detected gravitational wave (GW) signals. At the end of the LIGO-Virgo-KAGRA (LVK) Collaboration's third observing run, we have detected nearly 100 GW wave signals from compact binary coalescence (CBC). Over the past decade, scientific and data analysis techniques related to GW have made great progress, but GW science is still in its infancy. Looking ahead, in the 2030s, the third-generation ground-based GW detectors (such as Einstein Telescope and Cosmic Explorer) shall be completed and begin scientific observations, while LISA may open a new frequency band in space. These future detectors will open up new scientific opportunities while also bringing new challenges, requiring the development of new data analysis methods. In this thesis, we review the GW events up to the third observing run, reproduce these scientific results by analyzing public data, and verify the reliability of the results. Then, we discuss the issues that third-generation ground-based detectors will encounter. Due to their tenfold increase in sensitivity and lower low-frequency cutoff, CBC signals will overlap in the time-frequency domain, forming a foreground noise composed of GW signals. We discuss in detail its impact on matched filtering and possible strategies to reduce the impact. Subsequently, we discuss how to optimally combine data from third-generation ground-based GW detectors and the space-borne detector LISA to achieve multiband observations of stellar-mass binary black holes and intermediate-mass binary black holes. Our new coherent multiband method can analyze LISA signals with a signal-to-noise ratio (SNR) as low as 3, breaking through the long-standing observational bottleneck. Finally, we introduce a more flexible and efficient method for constructing a GW template bank and apply it to the detection of supermassive binary black hole signals in LISA mock data.