## Signal analysis in gravitational wave data

##### Doctoral thesis

##### Permanent lenke

https://hdl.handle.net/11250/3114461##### Utgivelsesdato

2024##### Metadata

Vis full innførsel##### Samlinger

- PhD theses (TN-IMF) [16]

##### Originalversjon

Signal analysis in gravitational wave data by Paolo Marcoccia, Stavanger : University of Stavanger, 2024 (PhD thesis UiS, no. 740)##### Sammendrag

In this thesis, I study the gravitational wave signals coming from compact objects on both present-stage, and future gravitational wave detectors.
The thesis is based on 3 articles, which will be presented after two brief introductory chapters, aimed to guide the reader through some of the main concepts and tools required in the analysis of the presented papers.
In the first article we adopt the Pearson Cross-Correlation analysis, to perform an agnostic search on real detector data of the first four LIGO gravitational wave detections. This work was motivated as a follow-up to some studies carried out by a group at the Niels-Bohr Institute. In their works, they tried to reproduce the detections claimed by the LIGO collaboration using matched filtering, and they discovered that the waveforms used by the LIGO collaboration in their subtractions were not optimal, as some of the signal remained buried in the detector noise after the subtraction. In the paper we used different waveforms, obtained through maximum-likelihood, and we demonstrate that the residual signal found in the noise was just a consequence of the choice of waveforms. Such signal, buried in the residual detector noise, is hence not a result of mismatching on the model but can be removed by using a more accurate waveform description. Furthermore, we show that the LIGO results can be reproduced with statistical significance even by using the Pearson cross-correlation method, even though with this approach the statistical significance will be slightly lower compared to the results obtained using matched filtering.
For the second article we moved to the case of simulated signals, coming from many events, on a future space-based detector datastream. To this extent we analyze the Stochastic Gravitational Wave Background predicted on the LISA detector, which is given by the superposition of all the weak unresolvable signals on the detector strain. We forecast the signal on the LISA detector strain by using the results coming from the latest LVK population inference paper, and produced catalogs representing a Stellar-Origin Black Hole population in our Universe. The Stochastic Gravitational Wave Background is then computed by adopting four different methods, that in order of complexity, range from a simple analytical evaluation to estimating the real detector strain after synthesizing a black hole population and iteratively subtracting all the resolvable sources. We find that, when the assumed SNR threshold is high enough to keep the number of resolvable sources small (∼ 10 over 4 years of observation), all the methods give results well in agreement with each other. This implies that, when working with LISA data, it is possible to use the fast analytical estimation for the stochastic noise component with a small loss of precision. On the other hand, the use of more complex methods like the iterative subtraction of a synthesized population, despite naturally requiring numerical cuts in the population generation phase due to its computational cost, can present both the value of the Stochastic Gravitational Wave Background amplitude as well as the resolvable sources predicted on the LISA strain at the same time. It can hence be useful when both these quantities need to be taken into account in a particular study.
We conclude on the third article by studying the synergy of multiple future gravitational wave detectors (both Earth-based and Space-based), in assessing the presence of secondary population channels in the detectors data stream. To this extent, we investigate the prospects of identifying potential Primordial Black Hole Binary populations over the astrophysical Stellar-Origin Black Hole Binary population of our Universe. We once again assume that our fiducial population follows the latest LVK GWTC-3 inference paper results, and we forecast our analysis on the next generation of gravitational wave detectors. We consider different possibilities both for the merger rate and mass function of the studied Primordial Black Hole subpopulations, and we perform our analysis by focusing on the signatures at higher redshifts than the current LVK detection horizon. We exploit the fact that the astrophysical black holes of our universe are supposed to follow a distribution as a function of redshift closely related to the Star Formation Rate, which is supposed to peak and then slowly die off. At distances beyond the peak of the stellar formation rate, the Stellar-Origin Binary Black Hole contribution will hence become negligible, whereas Primordial Black Hole models predict many sources and will dominate. We generally find that Earth-based and space-based detectors work synergistically, and the value of the Stochastic Gravitational Wave Background measured by LISA will generally be able to improve constraining the presence of additional sub-populations compared to the case when only Earth-based detector observations are considered.

##### Består av

Paper 1: Marcoccia, P., Fredriksson, F., Nielsen, A.B. & Nardini, G. (2020) Pearson cross-correlation in the first four black hole binary mergers, Journal of Cosmology and Astroparticle Physics, 2020(043), DOI 10.1088/1475-7516/2020/11/043Paper 2: Babak, S., Caprini, C., Figueroa, D.G., Karnesis, N., Marcoccia, P., Nardini, G., Pieroni, M., Ricciardone, A., Sesana, A. & Torrado, J. (2023) Stochastic gravitational wave background from stellar origin binary black holes in LISA. Journal of Cosmology and Astroparticle Physics, 2023. DOI 10.1088/1475-7516/2023/08/034

Paper 3: Marcoccia, P., Nardini, G. & Pieroni, M. Probing primordial black holes at high redshift with future gravitational wave detector. [Pre-print]

##### Utgiver

University of Stavanger, Norway##### Serie

PhD thesis UiS;;740