TABASCAL: Removing multi-satellite interference from radio interferometry observations

¹ Département de Physique Théorique and Center for Astroparticle Physics, Université de Genève, 24 quai Ernest Ansermet, 1211 Genève 4, Switzerland
² Institute for Machine Intelligence and Neural Discovery (MIND), University of the Witwatersrand, Johannesburg, South Africa
³ School of Computer Science and Applied Mathematics, University of the Witwatersrand, Johannesburg, South Africa
⁴ Department of Mathematics and Applied Mathematics, University of Cape Town, Rondebosch, Cape Town 7700, South Africa
⁵ South African Radio Astronomy Observatory, Liesbeek House Building, River Park, Gloucester Road, Mowbray, Cape Town 7700, South Africa
⁶ Centre for Radio Astronomy Techniques and Technologies, Department of Physics and Electronics, Rhodes University, PO Box 94, Makhanda 6140, South Africa

^★ Corresponding author: dr.chris.finlay@gmail.com

Received: 17 March 2025
Accepted: 22 July 2025

Abstract

In the first trajectory-based radio frequency interference (RFI) subtraction and calibration (TABASCAL) paper, we showed how to calibrate radio interferometers in the presence of RFI sources by simultaneously isolating the trajectories and signals of the RFI sources. In this paper, we show that we can accurately remove RFI (i.e. recover the astronomical signal) from simulated MeerKAT radio interferometry target data. We are able to do so for a single frequency channel, corrupted by up to nine simultaneous satellites, with average RFI amplitudes varying from weak to very strong (1–10³ Jy). Additionally, TABASCAL also manages to leverage the signal-to-noise ratio (S/N) of the RFI to phase-calibrate the astronomical signal. TABASCAL, effectively performs a suitably phased up fringe filter for each RFI source, which essentially allows for an ideal removal of RFI across all RFI strengths. As a result, TABASCAL is able to reach image noises equivalent to the uncorrupted, no-RFI, case. For larger RFI amplitudes, the resulting image noise is 10×–100× smaller than those from traditional RFI flagging methods such as AOFLAGGER. As a specific application, we show that point-source science with TABASCAL almost matches the no-RFI case with near perfect completeness for all RFI amplitudes. In contrast, the completeness of AOFLAGGER and idealised 3σ flagging drops below 40% for strong RFI amplitudes, where recovered flux errors are approximately 10×–100× worse than those from TABASCAL. Finally, we note that TABASCAL works for astronomical sources with both static and varying fluxes.