Improving the sensitivity of searches for gravitational waves from compact binary coalescences

The detection of gravitational waves from the coalescence of two compact objects has been brought to within touching distance by the construction and operation of a global network of laser-interferometer detectors. However, the amplitude of the radiation from these events is so low that direct detection will require the combined innovations of advanced interferometry and detector characterisation, along with powerful methods of extracting weak, but modelled, signals from the background detector noise. This work focuses on enhancing the probability of such detection through improved identi�cation of noise artefacts in the instrumental data, and improved signal processing and extraction. We begin with a recap of the theory of gravitational waves as derived from Einstein's theory of gravity, and the mechanisms that allow propagation of this radiation away from a source. We also catalogue a number of promising astrophysical progenitors, with a focus on compact binary coalescences. We detail the interactions between gravitational waves and an observer, and describe the layout of the large-scale laser interferometers that have been built to enable direct detection. A description of the operation of these detectors during the last science run is given, focusing on their stability and sensitivity, isolating a number of key instrumental noise mechanisms and how they a�ected astrophysical searches over the data. Additionally, we illustrate a new method to improve the identi�cation of seismic noise bursts, allowing their removal from search data, improving search sensitivity. The LIGO and Virgo gravitational-wave detectors operated as a network during the last joint science run. A summary is given of the analysis pipeline used to search for gravitational waves signals from compact binary coalescences using a coincidence-based method, including details of the results of that analysis. Details are also given of the pipeline used to search for gravitational waves associated with short, hard gamma-ray bursts, in which a new coherent method was tuned to search over the reduced parameter space constrained by the electromagnetic counterpart. Finally, we present a new pipeline adapting the coherent method to the blind, all-sky, all-time search, allowing for a more sensitive analysis, as demonstrated by direct comparison