Low-latency Statistical Data Quality in the Era of Multi-Messenger Astronomy
Author | : Patrick Godwin |
Publisher | : |
Total Pages | : |
Release | : 2020 |
ISBN-10 | : OCLC:1198447290 |
ISBN-13 | : |
Rating | : 4/5 ( Downloads) |
Download or read book Low-latency Statistical Data Quality in the Era of Multi-Messenger Astronomy written by Patrick Godwin and published by . This book was released on 2020 with total page pages. Available in PDF, EPUB and Kindle. Book excerpt: August 8, 2017 marked the dawn of multi-messenger astronomy with the detection of gravitational waves from a binary neutron star merger from LIGO and Virgo, GW170817, and its joint detection of a short gamma ray burst observed by Fermi- GBM and INTEGRAL. This prompted a massive follow-up campaign with more than 70 telescopes and observatories participating in the search for multiple cosmic messengers from electromagnetic radiation, gravitational waves, cosmic rays and neutrinos. Being able to send out prompt alerts from gravitational wave detections is of utmost importance to capture the onset of electromagnetic emission from compact binary mergers, and this requires the rapid validation of gravitational-wave candidate events. While both LIGO-Hanford and LIGO-Livingston were operating at the time of the detection, a non-Gaussian noise transient, or glitch, coincided with the binary neutron star merger in LIGO-Livingston, causing the candidate event to be identified in low latency in only a single detector by the GstLAL analysis, a matched filter gravitational-wave search. The presence of non-Gaussian noise caused issues in the initial detection of the gravitational-wave signal, and signifies a need to provide rapid data quality information that can identify non-Gaussian noise transients and limit their impact in low-latency gravitational-wave searches. This dissertation focuses on methods to provide near-real-time data quality information and its incorporation into gravitational-wave searches. This includes a method to extract non-stationary noise features and using these features to perform real-time statistical inference of non-stationary noise in gravitational-wave data. Finally, I describe a procedure to incorporate statistical data quality information into the GstLAL search analysis, providing a way to autonomously folding in data quality information using the detector's auxiliary state for a wide set of glitch classes.