Agnostic method to detect low energetic signals nearby a gravitational wave transient from a binary black hole system

The first detection of a gravitational wave (GW) enabled our observation of the Universe through a revolutionary messenger and unveiled phenomena that are occurring in a range of very strong gravitational fields and relativistic velocities. These physical regimes, previously inaccessible to humankind, can now be studied. In particular, the discoveries of an unexpected population of stellar-mass binary black holes (BBH), and unexpected masses for binary neutron star (BNS) components have both pointed to new astrophysics, and to unprecedented tests of the general relativity theory. This thesis focuses on the development of a new method of gravitational wave data analysis, aiming to investigate weak features in the proximity to well-identified BBH merger signals. The method is based on a dedicated version of coherentWaveBurst (cWB), an unmodelled gravitational waves transient search algorithm, developed in the LIGO Scientific Collaboration (LSC) and Virgo Collaboration and widely used on LIGO-Virgo-KAGRA (LVK) data. CoherentWaveBurst relies on the coherent detection of an excess of energy inside the combined data of all the gravitational waves detectors inside the detectors network. Such excess of energy must pass several internal thresholds of the pipeline to be accepted as a possible gravitational wave candidate and these thresholds evaluate not only the strength of the signal with respect to the background noise but also how balanced is the energy distribution among the detectors of the network, its coherence, as well as other quantities whose purpose is to rule out possible outliers due to the presence of non-stationary noise. To develop such a method, it was decided to adopt as science case the search for echoes. In literature, it has been proposed that the gravitational radiation generated from a binary compact objects (CBCs) coalescence might display exotic characteristics if compared to the predicted one generated by black hole-black hole (BH-BH), neutron star-neutron star (NS-NS), or neutron star-black hole (NS-BH) binaries which are, for now, the only detected emitters of gravitational waves. Such differences arise from the proposal that the involved compact objects (COs) of the binary are not standard black holes but instead black hole mimickers called exotic compact objects (ECOs). If this is the case the gravitational wave signal generated from such a binary would display repeated gravitational wave pulses, of widely uncertain morphology, after the merger-ringdown phase of the gravitational signal. These repeated gravitational wave pulses are called echoes, one class of low energetic signals whose presence inside gravitational wave data, this new algorithm is searching for. The proposed data analysis methodology searching for echoes is agnostic over the properties of the predicted gravitational wave pulses emitted by an ECO binary. Indeed, the variety of theoretical alternatives to black holes is not converging over a well-defined post-merger-ringdown signal, each model has its own properties and characteristic features. Therefore, the possibility to investigate the morphological features of possible outliers in the post-merger phase of detected GW signals is fundamental in the process of inferring their nature. Having their morphology recovered without priors makes the proposed search more general than the variety of theoretical models of echoes. This procedure is tested over real data from past LIGO-Virgo observing runs (O1, O2, and O3), and the capability of the search in estimating the main morphological parameters of echoes, such as their arrival time, mean frequency, as well as the amplitude attenuation between subsequent pulses, is investigated. This work concludes that the current state-of-the-art methods and detectors find no evidence for echoes of any morphologies. Such a study extended to lower signal-to-noise ratio (SNR) the detectability of echoes associated with the public gravitational-wave transient catalog of BBH mergers released by the LIGO and Virgo Collaboration. It also sets best quantitative upper limits on the amplitude of low energy signals occurring after the merger-ringdown. To achieve these results, new post-processing tools are developed and optimised to detect and characterize possible energy excess inside a user-defined time window. This required the development of the code and to adapt the cWB infrastructure to the new working requirements which also involves a re-tuning of cWB itself. The optimization of the performances is based on off-source simulations for assessing the detection efficiency and false alarm probability of signal candidates.