Application of Particle Swarm Optimization in Gravitational Wave Detection

Bhaumik, Shubhagata (2019) Application of Particle Swarm Optimization in Gravitational Wave Detection. Masters thesis, Indian Institute of Science Education and Research Kolkata.

PDF (MS dissertation of Shubhagata Bhaumik (14MS077)) 14MS077.pdf - Submitted Version Restricted to Repository staff only Download (16MB)

Abstract

The detection of gravitational waves by LIGO in September, 2015 was one of the greatest and most exciting discoveries in the history of science. This discovery not only confirmed one of the predictions of Einstein’s General Relativity and further strengthened the theory, but it also provided the scientific community with a new means to understand and study the Universe. For example, before this discovery, there were no direct evidences of the existence of black holes since they do not emit light. But after this discovery was made, using the information obtained from the detected gravitational waves, one could not only establish the existence of black holes but also find out the masses, spins, sky position and other parameters of the binary black hole system that emitted the gravitational waves. In recent years, this has been a subject of great interest and several researchers of the LIGO consortium are pursuing research in this area. Unfortunately, one of the major problems that are faced by LIGO and other interferometers in gravitational wave (GW) detection is due to the fact that GWs that reach Earth have very small amplitude. The strain caused by GWs as they pass through matter on Earth is of the order 10⁻²¹. Because of this small amplitude, many a times the instrumental noise masks the presence of a GW waveform that is present in the data. In this thesis, we study various techniques used by LIGO to search for GWs within the noisy data obtained from the interferometer, and estimate the parameters of the binary black hole merger that emitted the gravitational wave. We use one of these techniques, namely, the matched filter, to find more efficient methods of searching for a GW hidden within the data. In particular, this thesis focuses on the application of the Particle Swarm Optimization (PSO) algorithm in searching for GWs in data. Other new algorithms with modifications to the PSO algorithm are also developed during the course of this thesis. After testing these algorithms on model signals without noise, these algorithms are applied to data with noise incorporated from the O2 run that has been released by LIGO. The working and performance of these algorithms on real data from the O2 run as well as constructed data is compared and analyzed.

Actions (login required)

View Item