Learned radio interferometric imaging for varying visibility coverage

RAS Techniques and Instruments, vol. 4

Abstract

ABSTRACT With the next generation of interferometric telescopes, such as the Square Kilometre Array (SKA), the need for highly computationally efficient reconstruction techniques is particularly acute. The challenge in designing learned, data-driven reconstruction techniques for radio interferometry is that they need to be agnostic to the varying visibility coverages of the telescope, since these are different for each observation. Because of this, learned post-processing or learned unrolled iterative reconstruction methods must typically be retrained for each specific observation, amounting to a large computational overhead. In this work we develop learned post-processing and unrolled iterative methods for varying visibility coverages, proposing training strategies to make these methods agnostic to variations in visibility coverage with minimal to no fine-tuning. Learned post-processing techniques are heavily dependent on the prior information encoded in training data and generalize poorly to other visibility coverages. In contrast, unrolled iterative methods, which include the telescope measurement operator inside the network, achieve good reconstruction quality and computation time, generalizing well to other coverages and require little to no fine-tuning. Furthermore, they generalize well to more realistic radio observations and are able to reconstruct images with with a larger dynamic range than the training set.

Authors 3

1. Matthijs Mars corresponding Aachen Leiden Observatory

   Leiden University · University College London · Leiden Observatory

   Affiliation as printed

   Leiden Observatory, Leiden University , NL-2333 CC Leiden ,

   Mullard Space Science Laboratory (MSSL), University College London (UCL) , Dorking RH5 6NT ,

   Leiden Observatory, Leiden University, NL-2333 CC Leiden, the Netherlands

   Mullard Space Science Laboratory (MSSL), University College London (UCL), Dorking RH5 6NT, UK

2. Marta M. Betcke

   University College London

   Affiliation as printed

   University College London (UCL) Department of Computer Science, , London WC1E 6BT ,

   Department of Computer Science, University College London (UCL), London WC1E 6BT, UK

3. Jason D. McEwen corresponding

   University College London · The Alan Turing Institute

   Affiliation as printed

   Alan Turing Institute , London NW1 2DB ,

   Mullard Space Science Laboratory (MSSL), University College London (UCL) , Dorking RH5 6NT ,

   Alan Turing Institute, London NW1 2DB, UK

   Mullard Space Science Laboratory (MSSL), University College London (UCL), Dorking RH5 6NT, UK