Multi-Step Reconstruction of Radio-Interferometric Images

TL;DR

This paper introduces a multi-step radio-interferometric image reconstruction method that partitions data by baseline length, enabling efficient processing of large-scale astronomical data with comparable image quality and computational cost.

Contribution

It presents a novel multi-step reconstruction approach that allows baseline-based data partitioning, improving flexibility and parallelization in processing large radio-astronomical datasets.

Findings

Enables baseline-based data partitioning without significant drawbacks.

Maintains comparable image quality to traditional methods processing all baselines.

Operates efficiently within existing imaging pipelines.

Abstract

The advent of large aperture arrays, such as the currently under construction Square Kilometer Array (SKA), allows for observing the universe in the radio-spectrum at unprecedented resolution and sensitivity. However, these telescopes produce data on the scale of exabytes, introducing a slew of hardware and software design challenges. This paper proposes a multi-step image reconstruction method that allows for partitioning visibility data by baseline length. This enables more flexible data distribution and parallelization, aiding in processing radio-astronomical observations within given constraints. The multi-step reconstruction is separated into two-steps, first reconstructing a low-resolution image with only short-baseline visibilities, and then using this image together with the long-baseline visibilities to reconstruct the full-resolution image. The proposed method only operates in the minor-cycle, and can be easily integrated into existing imaging pipelines. We show that our proposed method allows for partitioning visibilities by baseline without introducing significant additional drawbacks, having roughly the same computational cost and producing images of comparable quality to a method in the same framework that processes all baselines simultaneously.