Fast search method for Low Frequency Continuous Gravitational Wave Signals

Abstract

The detection of GW150914 signal from a binary black hole merger marked the beginning of observational gravitational wave astronomy. This has provided us with confidence and insights towards already well-known theory of general relativity. Though the signals from mergers are strong enough to be detected, spinning asymmetric neutron stars are predicted to have quasi monochromatic gravitational waves at very low strain amplitudes. The signals are also frequency modulated due to motion of the detector along with the Earth’s rotation and its orbital motion around the Sun. The demodulation requires some correction which depends on the signal parameters. The location of the stars are not known beforehand for most of the cases as most of them don’t emit any detectable signals in the electromagnetic spectrum. For example, astrophysically there should be around ∼ 109 neutron stars in our galaxy while only ∼ 3000 of them are detected in the electromagnetic spectrum as pulsars. Gravitational wave signals are not yet detected. For an all-sky search, the demodulation is required to be performed over large number of parameter space patches. This makes the search computationally very expensive. In this project, we have explored a method which aims to mitigate the doppler modulations to reduce the number of patches in the parameter space that is required to be searched over. We successfully show that the method is computationally very inexpensive for many relevant cases when compared to other fullcoherent search methods. We also show that our method is more sensitive than the earlier methods for low frequency signals given the LIGO detector sensitivity in the O4 run.