Extreme-scale precision imaging in radio astronomy

The EIRA (Extreme-Scale Precision Imaging in Radio Astronomy) collaboration will focus on radio astronomy, which uses radio telescopes to collect data. This allows observation of the sky with antennae arrays at otherwise inaccessible angular resolutions and sensitivities. Algorithms being developed at Heriot-Watt University will address the challenges of building images from these incomplete linear data sets.

Extreme data sets 

In particular, next-generation radio telescopes such as the Square Kilometre Array (SKA), which is capable of imaging the sky at much higher resolution than current instruments, will generate Exabyte data volumes. This data can be used to construct image cubes (consisting of images across a wide number of frequencies) that will exhibit rich structure and reach sizes between 1 Terabyte (TB) and 1 Petabyte (PB). Imaging algorithms, and the applications that implement them, need to parallelise efficiently and scale up to large amounts of computational resources to handle these extreme data sets.

Better, more reliable images

The EIRA project will extend imaging algorithms to significantly improve the achievable resolution and dynamic range of the constructed images to include the calibration functionality that is needed to correct for instrumental and ionospheric effects on the data, as well as providing methodology for quantifying the uncertainty around the constructed image. 

Together these new features will enable high-resolution images with uncertainty quantification information, giving researchers confidence in the features that are present in the images created by the programs that EIRA will develop.

Parallelisation of algorithms

The parallelisation of these algorithms is one of the key aspects that EPCC is focusing on, and we will be building on an existing collaboration where the Puri-Psi and Psi codes were developed to provide wide-band parallel imaging functionality. 

Extending these applications to include calibration and uncertainty quantification, and developing novel domain decomposition techniques for this new set of functionality, will enable Puri-Psi and Psi to be utilised by the radio astronomy community for a wide range of imaging applications.

Medical imaging

Radio astronomy imaging algorithms are also applicable to medical imaging, where magnetic resonance imaging (MRI) and ultrasound systems utilise a similar approach to data collection and image construction as radio telescopes. We will extend our applications into a proof-of-concept implementation for 3D MRI and ultrasound devices once the applications have been tested in radio astronomy. 

Image shows the Cygnus A radio galaxy constructed from VLA data provided by NRAO using Puri-Psi. By Adrian Jackson.