Characterising Gravitational Wave Signal Detection Pipe-line

Nandi, Debasmita (2022) Characterising Gravitational Wave Signal Detection Pipe-line. Masters thesis, Indian Institute of Science Education and Research Kolkata.

Text (MS dissertation of Debasmita Nandi (17MS191)) 17MS191_Thesis_file.pdf - Submitted Version Restricted to Repository staff only Download (2MB)

Abstract

The General Theory of Relativity proposed by Einstein in 1915 is considered to be the most successful theory of gravity till now. In this theory gravity is explained as a curvature of the spacetime fabric. One of the predictions of this theory is the existence of the gravitational waves. It claims that the information about any change in the local gravitational field is transmitted to other points of the universe as ripples of the spacetime. These ripples are nothing but gravitational waves. Einstein predicted this in 1915 and after 100 years gravitational waves have been detected. For the first time scientists were able to detect the gravitational waves emitted by the merging of two black holes in the year 2015. This marks the beginning of a new era of gravitational wave astronomy. At present there are two LIGO detectors, one VIRGO and one KAGRA detector which are being used to detect the gravitational wave signals. The weakness of these signals makes it challenging to retrieve them from a comparable noise background. There are some search pipelines which use di↵erent algorithms to detect the gravitational wave signals. In this work we discuss the algorithm which is being used at present to analyse the gravitational wave data. In this method a bank of template waveform is used and each template from this bank is matched with the data. The one which gives maximum match is considered to be the solution. But this method is computationally costly and it slows down the whole process of detection. We propose a new algorithm called particle swarm optimization(PSO) which is much faster than the existing template bank method. In this algorithm a single swarm of particles or multiple such swarms are introduced in the parameter space. The particles explore the parameter space and find the optimal solution. We have tested PSO both with the simulated Gaussian noise and the real noise recorded during the third observation run of LIGO. The performance of PSO is compared in both the cases. We have tried to determine exactly how many particles and swarms are required to detect a gravitational wave signal with reasonable accuracy. It is evident from our results that PSO can perform satisfactorily with lower number of particles which in turn lowers the number of matchfiltering operations and thus make the process faster.

Actions (login required)

View Item