This pipeline aims to automatically measure the autocorrelation function for CHIME baseband FRBs without the need for humans to interfere, e.g. identifying the FRB on-pulse range by hand. It is written in an object-oriented manner, so that each key step in the pipeline uses a class.
The full pipeline is FitScintPipeline.FitScintPipeline.run()
.
Given FRB ID, DM, and the output folder path, the pipeline flow works as explained below:
WaterfallLoader
reads in the raw baseband data, dedisperse the raw data, and mask bad channels and RFISpectraCalculator
measures the on-pulse range, upchannelizes the data in the on-pulse range, and upchannelizes random off-pulse rangesSpectraFitter
fits smooth spectra for both on-pulse region and off-pulse rangesACFCalculator
measures the autocorrelation functions for the on-pulse and off-pulse regionsACFFitter
fits Lorenzian to the measured autocorrelation functions
Each step above can be taken out individually for testing.