Daily Image

20-12-2018
PreviousNext
Click here or on the picture for a full size image.

PhD Defence Shahrzad Naghibzadeh

Submitter: Albert-Jan Boonstra
Description: On Friday December 7th, Shahrzad Naghibzadeh successfully defended her PhD thesis Image Formation for Future Radio Telescopes at TUDelft. Starting in 2014, Dr Naghibzadeh started her work on image formation in the NWO BBBD DRIFT project at TUDelft. This was conducted in the context of the ASTRON-IBM DOME project, and supported by the Province of Drenthe and the Ministry of Economic Affairs.

The advent of exa-scale next generation radio telescopes poses challenges in data handling, requiring data to be processed quasi-realtime. Traditional pixel-based imaging algorithms such as CLEAN are not easily scalable to very large images especially when combined with high spatial resolution requirements. Also the processing load for traditional techniques becomes an issue. A related challenge is that radio astronomy imaging in most cases is an ill-posed problem, meaning that a single unique solution is not guaranteed.

To tackle this, dr Naghibzadeh applied several regularization approaches, and split the field of view into several regions for different source densities. Using efficient iterative inversion methods, including the so-called Krylov subspaces (a concept introduced in 1931), it was shown that the new approach (PRIFIRA) is computationally less demanding than traditional pixel-based approaches, and a good candidate for the snapshot imaging pipeline of the next generation radio telescopes.

This is not the end of this line of research, indeed, further testing on large data sets and joint calibration and imaging studies with PRIFIRA are logical next steps. But that is to be defined in the context of a new project.
Copyright: ASTRON
 
  Follow us on Twitter
Please feel free to submit an image using the Submit page.