June 16, 2017 | Written by: Rhonda Edwards and Graham Mackintosh
Categorized: IBM Cloud | Space Exploration
Share this post:
Earlier this year, the world took note when an international team of scientists discovered that three of seven Earth-size planets orbiting the nearby star TRAPPIST-1 were in the habitable zone. Solar systems with this many planets are a rare find in the galaxy. Even rarer, these planets are similar in size to Earth and could have water on their surfaces to potentially support life. Needless to say this cosmic finding has been exciting for researchers and citizens alike. But for our team of researchers and data scientists at IBM and the SETI Institute, this is just the moment we’ve been waiting for. With the IBM Cloud and the IBM Data Science Experience, SETI is able to expand its search and experiment with new observation methods to support the search for life on TRAPPIST-1.
Searching for Extra-Terrestrial Intelligence in Lines of Data
For more than 30 years, the SETI Institute has been a pioneer in the search for extra-terrestrial intelligence. A pivotal moment was the construction of the Allen Telescope Array (ATA) in 2007 – a radio telescope array that has been scanning the skies for interstellar radio waves ever since. The SETI Institute’s systems analyze and capture approximately 4.5 terabytes of data from the ATA every hour, resulting in the capture of more than 168 million rows of data in SignalDB, and more than 20 million complex-altitude files in the last 10 years. Due to the enormous size of these datasets, there is a wealth of data that has never been fully explored.
SETI scientists have long wanted to expand the use of the ATA to detect much more subtle wide band radio signals that may not have been intended for us at all, but they have lacked the means to capture, store and process all the terabytes of data needed to enable these more complex detection algorithms. Recent developments in technology are accelerating the speed of SETI searches, and according to SETI Institute researcher Dr. Seth Shostak, it is hardly fantastic to suggest that we could find evidence of cosmic company within a few decades.
A new observational technique, often called “eavesdropping,” is now being performed by SETI researchers due to increased storage and compute capacity made available through a partnership with IBM, and it is moving the SETI Institute forward into new scientific territory. Using Deep Learning tools and data science capabilities on the IBM Cloud, SETI researchers can tap into this more expanded use of the ATA and have the ability to observe a much wider array of signal types.
Enter TRAPPIST 1
At a mere 40 light years away from Earth, the newly discovered TRAPPIST 1 system presents a petri dish of possibilities in discovering new life. When observing the planets of the red dwarf star from Earth, the ATA telescope array can be focused to specifically look for communications that might transpire between them, much as we communicate with the space probes that we send to other the planets in our own solar system.
Using this new “eavesdropping” technique, the SETI Institute will be able to conduct several observations when two planets within the TRAPPIST-1 system come into alignment with Earth. During these observations, several terabytes of raw telescope data will be uploaded and searched using a high performance IBM Cloud system and researchers will be experimenting with new methods to discover symbolic patterns in the radio signals – patterns that might indicate some sort of language structure. You can follow along on these observations here. With these new capabilities, the likelihood of discovery is closer than ever before.
Separating the Signal from the Noise
Separating the Signal from the Noise is the mission for this project. Detecting that faint bit of transmission that might prove we are not alone is the goal. In the meantime, we continue to improve the technology that makes it possible.
This challenge of “weak signal to noise ratios” is actually common to nearly every industry and business function. Moreover, it is usually the weak signals in business that harbor the most interesting insight. Early indications of a new retail trend, hidden in the flood of mainstream purchasing patterns, is a weak signal. A few simple tweets that are geographically and temporally clustered might signal the start of a potential pandemic. The subtle signs of fraud, intentionally distributed across multiple transactions to avoid detection, carries a lot of similarities to the patterns of wide band radio signals, spread across a broad range of radio frequencies.
“Are we alone?” This is the sort of big question that IBM has always been drawn to because we know that tackling grand challenges opens up new capabilities and opportunities in other domains. What signals will you find in your noise now that you have the power of AI and data analytics on the cloud to help you?