Physics Master of Science Project Presentation by Bhaskar Verma
Abstract:
In the emerging field of gravitational wave astronomy, the data collected by gravitational-wave (GW) observatories is key to understanding the universe. However, in addition to astrophysical signals, the data consists of non-stationary detector noise and transient bursts of noise known as glitches. These glitches impact the ability to both observe and characterize incoming gravitational-wave signals. Thus, it is imperative that we study these glitch populations to improve our sensitivity to real signals and provide feedback to instrumentalists. Current glitch mitigation pipelines use glitch spectrogram images, which have been used to train many state-of-the-art glitch analysis tools. While this approach has proven to be effective, many aspects of the glitch, such as phase information, short-glitch events, and time localization, are lost. Due to these limitations of frequency-domain analysis, there is a need for glitch analysis tools that operate in the time domain. In this work, we present the first large- scale glitch time-domain model reconstruction analysis on glitch data from LIGO’s third observation run. We introduce a machine-learning based tool to assess the quality of glitch time-domain reconstructions by utilizing non-Gaussianity tests to analyze glitch residuals and enabling the optimization of time-series models for various LIGO glitch classes. Using this framework, we demonstrate how large-scale time-domain datasets of real, noise-free detector glitches can be rapidly produced and assessed, paving the way for improved glitch population studies and future developments in classification and simulation tools.
Advisor:
Dr. Sarah Caudill, Department of Physics (scaudill@umassd.edu)
Note:
All PHY Graduate Students are encouraged to attend.
SENG 201