Fast radio bursts or FRBs are brief but powerful pulses of energy from distant parts of the cosmos. The first one was spotted in 2007, and we're finding more of them all the time.

While astronomers have recently made some exciting progress in tracing FRBs, we just don't know exactly what these signals are, or how they originate. They might be caused by black holes or neutron stars called magnetars, perhaps.

What's exciting about the detection by FAST is that this fast radio burst is a repeater. The burst is officially known as FRB 121102: first picked up in 2012 at the Arecibo Observatory in Puerto Rico, it's appeared several times since.

Researchers note that the signal has travelled around 3 billion light-years across the Universe to reach us.

FAST latched on to FRB 121102 on August 30, before recording dozens of later pulses (on one particular day, September 3, more than 20 pulses were detected). So, this looks like a particularly persistent FRB.

The 19-beam receiver on FAST is especially sensitive to radio signals, covering the 1.05-1.45 GHz frequency range, and that makes it perfect for keeping an eye on FRB 121102.

The more observations we can make of these FRBs, the better our chances of being able to work out exactly what they are. One idea is that FRBs are produced upon disintegration of the crusts of certain types of neutron stars.

Another hypothesis posits that different FRBs actually have different causes, which may explain why FRB 121102 repeats and others don't appear to do so. We are at least getting better at pinpointing where these mysterious bursts of electromagnetic radiation come from.

Now we can add the data gathered by FAST to our growing database of knowledge on these most intriguing of space phenomena. The team at the telescope has already been able to eliminate aircraft and satellite interference from their measurements.

"I just think it is so amazing that nature produces something like that," physicist Ziggy Pleunis of McGill University told ScienceAlert, after helping to detail eight new FRBs in a paper published last month.

"Also, I think that there is some very important information in that structure that we just have to figure out how to encode and it has been a lot of fun to try to figure out what exactly that is."