Master's Project
Template Lattices for a Cross-Correlation Search for Gravitational Waves from Scorpius X-1
Scorpius X-1 is a low-mass X-ray binary (LMXB) source that is rapidly spinning and is a promising source for continuous gravitational wave emission. As a neutron star with a low-mass companion, its high x-ray emission indicates a high accretion rate. Resulting nonaxisymmetric deformations can lead to gravitational wave emission at twice the rotation frequency of the star. This source is currently being observed by detectors at LIGO, Virgo, and KAGRA (LVK). The signal received depends on the parameters of the system, and the sensitivity of a search for the signal in the data depends on a use of accurate values for those parameters. The search becomes less sensitive if incorrect values are used.
The fractional loss of signal-to-noise ratio can be written as a distance metric on parameter space, which we divide into small enough sections that we can assume flat space and constant metric values. This allows us to place a lattice over sections of parameter space, where points on the lattice are templates that search the space for a signal at discrete points. This then becomes the sphere covering problem - we need to ensure that the entirety of the parameter space is covered, but the use of additional templates causes computing cost to increase.
I have worked to find four different ways to decrease the number of templates needed to cover the parameter space while maintaining search sensitivity. This is useful especially because our cross-correlation search method is tunable. Increasing the coherence time between pairs of data improves the sensitivity but increases the computing cost, mostly by requiring a higher density of templates. So any way to speed up the search, such as by placing the templates more efficiently so you need fewer of them at the same coherence time, allows a search to be performed at the same sensitivity with fewer computing resources, or a more sensitive search with the same computing resources. See the methods paper on arXiv!
These improvements to the lattice are now implemented in our analysis pipeline, being used on data from LIGO's third observing run.