Noisy environments: detecting signals in gravitational waves detectors

Signals buried in noisy environments appear in several contexts. Earth-based gravitational wave detectors of the LIGO-Virgo-KAGRA (LVK) collaboration are a perfect example of instruments dominated by noise. Developing proper analysis techniques is fundamental for detecting gravitational wave signals. Due to the complexity of these instruments, fitting a model for noise is an unfeasible task, and the properties of the underlying stochastic process are better inferred from the noise realization itself. Despite being a challenging task, several assumptions about noise properties allow us to characterize these systems in terms of wide-sense stationary processes. The Bayesian framework is suitable for studying such processes, connecting data analysis techniques with the theory of stochastic processes. We will explore this framework and the main ideas that made detecting these signals possible, and give a hint on how Bayesian principles such as Maximum Entropy can be used to marginalize our estimates over noise uncertainty.