Determination of Gravitational wave mass parameters using Particle Swarm Optimization and its modifications

Mandal, Ankit (2019) Determination of Gravitational wave mass parameters using Particle Swarm Optimization and its modifications. Masters thesis, Indian Institute of Science, Education and Research Kolkata.

PDF (MS dissertation of Ankit Mandal (14MS154)) 14MS154.pdf - Submitted Version Restricted to Repository staff only Download (9MB)

Abstract

Gravitational waves are in my opinion one of the most popular, important and interesting thing happening to modern Astronomy. Gravitational-wave is a new way of probing the universe, and knowing that we can detect them now means that new young researchers like us can join the related researches and contribute our bit from now and forever. It is like Galileo making the first telescope and being able to resolve a planet. There are in total of 13 parameters related to a Gravitational-wave. This report basically addresses the simple case of estimation of the two mass parameters involved, namely, the masses of the two coalescing binaries emitting the Gravitational waves. For this, we use three cases. First one is the case of the only signal with no noise, the second case involves Gaussian noise in addition to the signal, the third and the most realistic case being the inclusion og LIGO O2 noise along with the injected signal. The report discusses with primarily three algorithms, the Particle Swarm Optimization, Projection and Rotation algorithm, all which tries to converge to the masses present in the signal. In the report, we took a particular combination of masses and considered the corresponding waveform as the signal and used the waveforms generated from various combination of masses from a well-defined search space as the template. Each waveforms in the template were then cross-correlated with the signal. the cross-correlated values were then interpreted to allow the algorithms to converge to the injected masses in the signal. Then, we show the results of each algorithm with their inferences. These inferences were then concluded with an eye on future improvements.

Actions (login required)

View Item