http://arxiv.org/abs/1906.09670
We present a machine learning (ML) based method for automated detection of Gamma-Ray Bursts (GRBs) from the AstroSat CZTI data. We make use of density-based spatial clustering to detect excess power and carry out an unsupervised hierarchical clustering across all such events to identify the different categories of light curves present in the data. This representation helps in understanding the sensitivity of the instrument to the various GRB populations and identifies the major non-astrophysical noise artifacts present in the data. We make use of dynamic time wrapping (DTW) to carry out template matching to ensure the morphological similarity of the detected events with that of known typical GRB light curve. DTW alleviates the need for a dense template repository often required in matched filtering like searches, and the use of a similarity metric facilitates outlier detection suitable for capturing previously unmodeled events. Using the pipeline, we detect 35 new GRB events and briefly report their characteristics in this paper. Augmenting the existing data analysis pipeline with ML capabilities enables the instrument to quickly respond to alerts received from observatories such as the gravitational wave detectors and carry out robust follow up studies without the need for an onboard classification facility.
S. Abraham, N. Mukund, A. Vibhute, et. al.
Tue, 25 Jun 19
19/68
Comments: 6 pages, 6 figures
You must be logged in to post a comment.