Chirps vs. hums: Statistical challenges of studying gravitational-wave signals from coalescing binary systems and rotating neutron stars

Metro Toronto Convention Centre 

Gravitational waves (GWs) come in many shapes and forms. Different types of signals present unique statistical challenges, requiring specialized techniques to extract a signal from noisy detector data. The GW signals observed thus far are transient, short-duration signals originating from compact binary coalescences (CBCs), i.e. mergers between black holes and neutron stars. Bayesian inference techniques have proven to be extremely powerful tools for extracting the source parameters of these merging binaries. Another class of GWs are the as-yet undetected "continuous waves" (CWs), created by rapidly rotating neutron stars. Unlike CBC signals, CWs are long-lived narrowband signals with amplitude orders of magnitude below the noise in the detector, requiring an entirely different set of search techniques. Adding to the difficulty of CW searches is the possibility that a signal may wander in complicated ways, which will need to be tracked over long periods of time. In this talk I will discuss the contrasting statistical challenges one faces when trying to study these two classes of signals, as well as the techniques employed by the GW community to tackle them.