Webinar: Towards efficient, RFI-robust interferometric imaging

Towards efficient, RFI-robust interferometric imaging: dynamic antenna switching, EM models and algorithms

Modern radio telescopes like the Square Kilometre Array (SKA) offer exceptional sensitivity, but this comes with major challenges in terms of processing massive amounts of data and mitigating radio frequency interference (RFI). To reduce the computational load while maintaining imaging quality, we propose a subnet switching strategy based on variants of the EM algorithm. Different antenna sub-networks are used, each of which can be designed for specific properties. These sub-networks are obtained either by optimising minimum estimation bounds or through learning. RFI is robustly modelled using a Student’s t-distribution. Our models and methods are validated by simulations and tested on real observational data.