A Bayesian approach to RFI mitigation

A Bayesian approach to RFI mitigation by S. A. K. Leeney et al. on Tuesday 29 November Radio Frequency Interference (RFI) is an endemic problem in radio astronomy. Information in contaminated frequency channels is lost and it can lead to significant systematic error if not properly modelled. In this paper we propose a RFI mitigation methodology that takes a Bayesian approach, where contaminated data is both flagged and managed as part of a single step fitting process. To the authors knowledge, our approach is first-of-its-kind. We provide a proof-of-concept of our methods which are also tested and shown to be effective on both a simple toy model and when incorporated into the Bayesian data analysis pipeline for REACH (a modern cosmological radio experiment). arXiv: http://arxiv.org/abs/http://arxiv.org/abs/2211.15448v1