Big Data and Big Data: Searching for Interference in Electromagnetic Signals from Extrasolar Planets with a Large Telescope
However, Mr. Gertner writes, “rather than looking for direct calls to Earth, telescopes now sweep the sky, searching billions of frequencies simultaneously, for electronic signals whose origins can’t be explained by celestial phenomena.”
What scientists are most excited about is the prospect of other planets’ civilizations being able to create the same “telltale chemical and electromagnetic signs,” or, as they are now called, “technosignatures.”
“The biggest challenge for us in looking for SETI signals is not at this point getting the data,” says Sofia Sheikh, an astronomer at the SETI Institute. The challenge is to differentiate signals from technology in other places in the world and from those on Earth.
The problem of big data is relatively new for SETI. For decades, the field was constrained by having hardly any data at all. Astronomer Frank Drake pioneered SETI in 1960, when he pointed a telescope in Green Bank, West Virginia, towards two stars and listened for radio transmissions. The SETI searches were limited to a small number of stars.
These searches result in a huge amount of data, including false positives from interference from mobile phones, gps and other modern life things.
It is not practical to manually go through millions of observations. A common alternative approach is to use algorithms that look for signals matching what astronomers think alien beacons could look like. There are potentially interesting signals that are a little different from what is expected.
Enter the world of machine learning. Machine-learning algorithms are trained on large amounts of data and can learn to recognize features that are characteristic of Earthly interference, making them very good at filtering out the noise.
Sampling Green Bank Stars: Searching for the Candidate Stars at the BNL Optical Gravitational Lensing Telescope
Peter Ma is a mathematician and physicist at the University of Toronto. He says that they can’t always be sure of what will send them.
Ma and his co-workers looked at the observations of 820 stars made with the Robert C. Byrd Green Bank Telescope. They built machine- learning software to analyse the data, which netted nearly three million signals of interest, but discarded most as Earth based interference. Ma then manually reviewed around 20,000 signals and narrowed them down to 8 intriguing candidates.