NASA probe typically relies on human-controlled radio systems to interact with Earth. As the group of space data increases, NASA resembles cognitive radio, the infusion of artificial intelligence into space interactions networks, to meet demand and increase efficiency.
“Modern space communications systems use complicated software to support science and exploration operations,” stated Janette C. Briones, principal researcher in the cognitive communication plan at NASA’s Glenn Research Center in Cleveland, Ohio. “By implementing artificial intelligence and machine learning, spacecraft command these systems seamlessly, making real-time determinations without awaiting instruction.”
To learn cognitive radio, it’s easiest, to begin with, ground-based applications. In the U.S., the Federal Communications Commission designates parts of the electromagnetic spectrum used for communications to multiple users. What happens when no fixtures are left? How could a tool access the electromagnetic spectrum when all the taps are taken?
Software-defined radios like cognitive radio use artificial intelligence to employ underutilized divisions of the electromagnetic spectrum without human interference. These “white spaces” are currently unemployed, but already licensed, segments of the spectrum. The FCC allows a cognitive radio to use the frequency while unencumbered by its initial user until the user becomes effective again.
Regarding our metaphorical watering hole, cognitive radio draws on the water that would otherwise be wasted. The cognitive radio can use various “faucets,” no matter the frequency of that “faucet.” When a licensed device stops utilizing its frequency, cognitive radio maps from that customer’s “faucet” until the original user needs it again. Cognitive radio shifts from one white space to another, using electromagnetic taps as they become accessible.
“The recent improvement of cognitive technologies is a new thrust in the structure of communications systems,” said Briones. “We visualize these technologies will make our communications networks more effective and flexible for missions exploring the depths of space. By combining artificial intelligence and cognitive radios into our networks, we will enhance the efficiency, autonomy and authenticity of space communications systems.”
For NASA, the space conditions present unique hurdles that cognitive radio could decrease. Space weather, electromagnetic radiation released by the sun and other celestial bodies, fills space with noise that can interfere specific frequencies.
“Glenn Research Center is testing in creating cognitive radio applications able of recognizing and accommodating to space climate,” said Rigoberto Roche, a NASA cognitive engine improvement lead at Glenn. “They would broadcast outside the range of the interference or cancel distortions within reach using machine learning.”
In future, a NASA cognitive radio could also learn to close itself down momentarily to mitigate radiation loss during critical space weather phenomena. Adaptive radio software could avoid the harmful effects of space weather, increasing science and investigation data returns. A cognitive radio network could also recommend alternate data paths to the ground. These methods could prioritize and route data by multiple paths simultaneously to avoid interference. The cognitive radio’s artificial intelligence could also designate ground station downlinks just hours in advance, as opposed to weeks, leading to an adequate scheduling.
Also, the cognitive radio may make communications network operations more efficient by reducing the need for human interruption. An intelligent radio could accommodate late electromagnetic landscapes without human help and foretell common operational settings for different conditions, automating time-consuming methods earlier handled by humans.
The Space Communications and Navigation (SCaN) Testbed aboard the International Space Station provides engineers and researchers with devices to test cognitive radio in the space environment. The testbed houses three software-defined radios in expanding to a variety of antennas and equipment that can be configured from the earth or other spacecraft.
“The testbed retains us honest about the environment in orbit,” said Dave Chelmins, project manager for the SCaN Testbed and cognitive communications at Glenn. “While it can be assumed on the ground, there is a factor of unpredictability to space. The testbed gives this environment, a setting that needs the resiliency of technology improvements like cognitive radio.”
Chelmins, Rioche and Briones are just a few of many NASA engineers adapting cognitive radio technologies to space. As with most terrestrial technologies, cognitive techniques can be more challenging to achieve in space due to orbital mechanics, the electromagnetic circumstances and intercommunications with legacy instruments. In spite of these difficulties, integrating machine learning into existing space communications infrastructure will increase the efficiency, autonomy and authenticity of these systems.
The SCaN program office at NASA Headquarters in Washington presents strategic and programmatic oversight for communications infrastructure and development. Its investigation provides critical developments in connectivity from spacecraft to ground.