Decoding gravitational wave signals from compact binary coalescences: towards next- generation detectors

eng The detection of gravitational waves (GWs) has opened a new era in astrophysics. The first three observing runs of the Advanced LIGO and Advanced Virgo detectors (O1, O2, and O3) have led to the detection of over 90 GW events, all originating from compact object binaries. While the majority of these events have been identified as binary black hole (BBH) mergers, several binary neutron star (BNS) and neutron star–black hole (NSBH) mergers have also been observed. The final part of this thesis coincides with the ongoing fourth observing run (O4), which has already achieved a significantly higher detection rate, surpassing 200 detections halfway through the run. As detector sensitivities continue to improve, future observations by the LIGO–Virgo–KAGRA (LVK) Collaboration, along with upcoming facilities such as the Einstein Telescope (ET), Cosmic Explorer, and the LISA space mission, are expected to uncover a broader range of source properties. These observations will provide crucial insights into the strong-field regime of gravity, the population and evolution of compact objects, galaxy formation, cosmology, and fundamental tests of general relativity. Accurate and computationally efficient waveform models are essential for extracting the physical parameters of GW signals. As detector sensitivities improve, the demand for more accurate and versatile waveform templates becomes increasingly critical, see e. g. ~reviews on the challenges for ET and LISA. This thesis contributes to the advancement of waveform modeling in three directions that are essential for improving model accuracy and extending coverage across the BBH parameter space. Although current inspiral-merger-ringdown (IMR) waveform models can accurately describe BBHs on quasi-circular (QC) orbits with aligned spins, several key regions of the parameter space remain poorly explored. These include binaries with generic spin orientations (which lead to orbital precession), systems with high or extreme mass ratios, and binaries on eccentric orbits - all of which are astrophysically motivated and relevant for upcoming GW observatories. This thesis aims to address these limitations and support the development of physically robust, generic BBH waveform models that will be essential for GW data analysis with future third-generation ground-based detectors and the LISA space mission. First, we present a model for predicting the remnant properties (final mass and spin) of spin-precessing BBHs in QC orbits across a wide range of mass ratios. By combining insights from numerical relativity and black hole perturbation theory, this model helps bridge the gap between the extreme and comparable mass ratio regimes. This work lays groundwork for the development of waveform models that are applicable across the full mass-ratio spectrum, which is particularly relevant for LISA, as it will be sensitive to high- and extreme-mass-ratio inspirals in the millihertz frequency band. Second, we introduce IMRPhenomTEHM, a new IMR phenomenological waveform model for aligned-spin, eccentric BBHs. This model incorporates post-Newtonian corrections atop the QC baseline IMRPhenomTHM and significantly reduces computational cost compared to other state-of-the-art eccentric models. Its computational efficiency makes it well-suited for large-scale Bayesian inference studies and a strong candidate for standard parameter estimation pipelines in current and future detector networks. Motivated by the performance and accuracy of the model, in the last part of the thesis we apply IMRPhenomTEHM to a reanalysis of selected GW events, investigating the detectability of eccentricity and assessing the robustness of current QC assumptions in GW parameter estimation. Our BBH parameter estimation studies identify evidence for eccentricity in two GW events from O3, GW200129 and GW200208₂2, and reveal that the high-mass events GW190701 and GW190929 exhibit potentially eccentric features. Due to the proven efficiency of IMRPhenomTEHM, we reanalyze three NSBH events from GWTC-3, finding support for eccentricity in GW200105 and quasi-circular consistency for GW200115 and GW230529, while also highlighting the impact of systematic uncertainties related to signal duration, waveform model degeneracies, and data quality. These results underscore the importance of including eccentric waveform models in future analyses, particularly as the BBH population becomes increasingly diverse with upcoming observations. Accounting for these effects is crucial to mitigate potential biases in parameter estimation and to maximize the scientific output of next-generation GW detectors.