Brown astronomers create new technique to eliminate signals disrupting sensitive radio telescopes

PROVIDENCE, R.I. [Brown University] — Astronomers sifting through data from the Murchison Widefield Array, a radio telescope in Western Australia, found themselves confronting an unexpected mystery.

The telescope, which consists of 4,096 spider-like antennas designed to detect radio wave signals from more than 13 billion years ago, appeared to have stumbled upon something far more local: a television broadcast. This was puzzling, given that the telescope is located in a designated radio quiet zone, where the Australian government regulates signal levels from all radiocommunication equipment — including TV transmitters, Bluetooth devices, mobile phones and more — to minimize interference with the telescopes in the area. Even more perplexing, the television signal was streaking across the sky.

"It then hit us," said Jonathan Pober, a physicist at Brown University and the U.S. research lead for the Murchison Widefield Array project. "We said, 'I bet the signal is reflecting off an airplane.' We'd been seeing these signals for close to five years, and several people had suggested they were airplanes reflecting television broadcasts. We realized we might actually be able to confirm this theory for once.”

To do this, Pober enlisted Brown Ph.D. student Jade Ducharme for some astronomical detective work. The findings from the pair, published in Publications of the Astronomical Society of Australia, not only supported the airplane hypothesis but have now also provided astronomers with a new method to identify and filter out unwanted radio frequencies — a goal becoming increasingly important as Earth's skies grow noisier with the deployment of more satellites.

“Astronomy is facing an existential crisis," Pober said. “There is growing concern — and even some reports — that astronomers may soon be unable to carry out high-quality radio observations, as we know it, due to interference from satellite constellations. This is particularly challenging for telescopes like the Murchison Widefield Array, which observes the entire sky simultaneously. There's no way to point our telescopes away from satellites.”

Traditionally, when unwanted signals — known as radio frequency interference (RFI) — are detected in radio telescope data, those data are discarded as contaminated. This is because these signals are unpredictable, and without a clear model of their origin, it's nearly impossible to subtract them from the data, Ducharme explained.

"It ends up being insane amounts of data being thrown out to not have any part of the observation contaminated,” Ducharme said.

For Ducharme and Pober, the new study was about laying the framework to help solve this massive problem by developing a new method to trace RFIs from nearby objects. To do so, the pair combined two existing tracking techniques used in the field. The first, known as near-field corrections, adjusts the telescope to focus on objects closer to Earth, which normally cause interference. Telescopes are designed to look deep into space, but near-field corrections allow them to track nearby objects more accurately. The second technique, beamforming, sharpens the focus of an object by creating a more precise “beam” that pinpoints where the interference is coming from — in this case, bouncing off an airplane.

By combining the two methods, the researchers tracked the plane and analyzed how the reflected radio waves curved off of its surface. That allowed them to calculate that the airplane was flying at around 38,400 feet and moving at approximately 492 miles per hour. They also found that the RFI signal that bounced off the plane came from a frequency band associated with Australian digital TV Channel 7.

The team was unable to identify the specific flight due to incomplete publicly available flight logs, but Pober said the successful combination of the two techniques opens new doors for the field of radio astronomy.

“This is a key step toward making it possible to subtract human-made interference from the data," he said. “By accurately identifying and removing only the sources of interference, astronomers can preserve more of their observations, reduce frustrating data loss and increase the chances of making important discoveries.”

Next steps in the project involve trying to actually remove broadcast RFI signals from the data they looked at so that it remains useful for the MWA team. The scientists then hope to refine the method further and extend it to filter out interference from satellites and other space-based objects. The researchers note, however, that while the technique worked well for tracking airplanes, applying it to other sources of interference, like satellites, will be more challenging.

The study also highlights how rapidly the issue of RFI is growing. According to the United Nations Office for Outer Space Affairs, 11,330 satellites were orbiting Earth as of June 2023, a nearly 40% increase from January 2022. 

This satellite boom is only expected to expand in the coming decades, posing a major challenge for radio astronomy's ability to study phenomena such as black holes, galaxy formation and the origins of the universe. Science leaders have already taken some action. For example, since 2019, the National Science Foundation's National Radio Astronomy Observatory and SpaceX have been working together to develop real-time data-sharing systems to try to minimize satellites from interfering with telescope observations.

Still, the debate continues on whether any action will be enough as the world is increasingly filled with artificial signals. Some, like Pober, wonder if the best course of action is to escape the noise by going beyond it — and building radio telescopes on places like the Moon. 

"If we can't find a quiet sky on Earth, maybe Earth isn't the place to be," Pober said. "No matter what we do, we have no choice but to invest in better data analysis techniques to identify and remove human-generated interference.”