We are creating some awesome events for you. Kindly bear with us.

AI boosts detection of fast radio bursts in real-time

A PhD student from Australia’s Swinburne University of Technology has built an automated system that uses artificial intelligence (AI) in order to revolutionise the ability to detect and capture fast radio bursts (FRBs) in real-time, according to a recent press release.

About the initiative

FRBs are mysterious and powerful flashes of radio waves from space, which are thought to originate billions of light years from the Earth.

They last for only a few milliseconds, or a thousandth of a second, while their cause remains to be one of astronomy’s biggest puzzles.

Mr Wael Farah developed the FRB detection system and is the first person to discover FRBs in real-time with a fully automated, machine learning system.

The system he developed has already identified five bursts, which includes one of the most energetic ever detected, as well as the broadest.

His results have been published in the Monthly Notices of the Royal Astronomical Society.

Capturing fast radio bursts in real-time

The researcher was able to train the on-site computer, at the Molonglo Radio Observatory near Canberra, to recognise the signs and signatures of FRBs.

He also trained it to trigger an immediate capture of the finest details seen to date.

The bursts were detected within seconds of their arrival at the Molonglo Radio Telescope.

It was able to produce high quality data that allowed researchers from the University to study their structure accurately, and to gather clues about their origin.

The PhD student explained that his interest in FRBs comes from the fact they can potentially be used to study matter around and between galaxies that is otherwise almost impossible to see.

Additionally, it is fascinating to discover that a signal, which travelled halfway through the universe, reached telescopes on Earth after a journey of a few billion years.

It is fascinating to discover that the signal exhibits complex structure, like peaks separated by less than a millisecond.

A Molonglo Project Scientist shared that the student used machine learning on their high-performance computing cluster to detect and save FRBs from amongst millions of other radio events, such as mobile phones, lightning storms, and signals from the Sun and from pulsars.

An Australian Research Council Laureate Fellow and Project Leader, meanwhile, explained that Molonglo’s real-time detection system allows them to fully exploit its high time and frequency resolution and probe FRB properties that were previously unobtainable.

The five bursts were found as part of the UTMOST FRB search program, which is a joint collaboration between Swinburne and the University of Sydney.

The Molonglo telescope is owned by the University of Sydney.

World-first discoveries

In June, Swinburne astrophysicists Dr Adam Deller and Dr Ryan Shannon, from the Centre for Astrophysics and Supercomputing, were part of a team that determined the precise location of a one-off FRB, for the first time.

Dr Shannon also led the discovery of 20 FRBS in 2018, nearly doubling the known number of bursts at that time.

Send this to a friend