Radio signals discovered coming from star systems close to Earth.
Scientists searching for radio frequencies that indicate an unnatural origin have used machine learning, a subfield of AI that improves existing human technology, on programmes built to comb through data to find radio 'signals of interest' in space. Radio is a highly efficient way to send information over the vast distances between the stars as it quickly passes through the dust and gas that permeate space, and it does so at the speed of light.
Led by an undergraduate student at the University of Toronto, Peter Ma, along with researchers from the SETI Institute, Breakthrough Listen and scientific research institutions around the world, they have applied a deep learning technique to a previously studied dataset of nearby stars and uncovered eight previously unidentified signals of interest.
After searching through more than 100 million sources of data for 'technosignatures', they found 3 million signals of interest. From those 3 million the machine learning helped them to indicate which were false positives or caused by radiation interference. Narrowing the data down to 20,000 data sets, it was found that 8 of those were classified as 'signals of interest', all coming from stars between 30 and 90 light years from Earth.
Algorithms developed decades ago for the earliest digital computers can be outdated and inefficient. The goal was to apply new deep learning techniques to a classical search algorithm to yield faster, more accurate results. After running the new algorithm and manually re-examining the data to confirm the results, newly detected signals had several key characteristics:
The signals were narrow band, meaning they had narrow spectral width, on the order of just a few Hz. Signals caused by natural phenomena tend to be broadband.
The signals had non-zero drift rates, which means the signals had a slope. Such slopes could indicate a signal's origin had some relative acceleration with our receivers, hence not local to the radio observatory.
The signals appeared in ON-source observations and not in OFF-source observations. If a signal originates from a specific celestial source, it appears when we point our telescope toward the target and disappears when we look away. Human radio interference usually occurs in ON and OFF observations due to the source being close by.
It is described in a paper, ‘A deep-learning search for technosignatures from 820 nearby stars’, published in Nature Astronomy today.
- ASTRAL MAGAZINE