If aliens are out there, how do you imagine them to be? If common narratives in Hollywood are correct, they may exist as humanoid creatures — possessing advanced intelligence and sharing many of our anatomical features. Perhaps they are only microscopic in size, existing as bacteria- or amoeba-like creatures.
However you envision aliens, a team of scientists led by the University of Toronto has moved us one step forward in our quest to discover intelligent alien life. An undertaking that first began in the 1960s, the search for extraterrestrial intelligence has involved the use of powerful radio telescopes to parse through the stars and galaxies in our universe in pursuit of “technosignatures,” which are technologically generated signals from advanced extraterrestrial civilizations.
A source of difficulty in locating technosignatures lies in differentiating between signals generated by extraterrestrial life and human-generated interference. To address this challenge, the team of researchers, headed up by U of T undergraduate student Peter Ma and supervised by U of T project scientist Cherry Ng at the Dunlap Institute, applied a novel algorithm to distinguish between the desired signals and unwanted interference. Their machine learning algorithm allows for the rapid sorting of data that is taken in as well as detecting specific patterns.
The team’s innovative approach has already resulted in the discovery of eight new radio signals that had gone unidentified in earlier studies of the same data. These signals were emitted from five stars located between 30 and 90 light years away. Unlike interference, which is ever present, these signals were only detectable when viewing the stars. Furthermore, the signals varied in frequency over time to appear to have originated far from the telescope. For these reasons, they were considered to have possibly stemmed from extraterrestrial life.
Importantly, these two criteria are not surefire signs of extraterrestrial intelligence, as they can be met merely by chance. When attempting to replicate these findings using the Green Bank Telescope in West Virginia, the same patterns suggesting extraterrestrial life were not found. Nevertheless, the team’s algorithm has immense potential to accelerate our search for alien civilizations. Their findings, which were published last month in Nature Astronomy, will enable obscure patterns to be more easily detected in vast sets of data.
Looking ahead, the team hopes to expand their algorithm to more datasets and observatories, taking our searching capabilities from hundreds of stars to millions.
