Researchers with the U.S. Department of Commerce National Institute of Standards and Technology built computer hardware that resembles the human brain but need superconducting electronics to keep it from grinding to a halt.
The neuromorphic devices have already demonstrated considerable promise, but their complexity and speed were constrained by the usage of traditional digital electronics, according to an Oct. 7 NIST news release. The signals between the separate components of larger and more complicated processors grow backed up like traffic on a gridlocked freeway, slowing computation to a halt.
“We could use what we’ve demonstrated here to solve computational problems, but the scale would be limited,” NIST project leader Jeff Shainline said in the release. “Our next goal is to combine this advance in superconducting electronics with semiconductor light sources. That will allow us to achieve communication between many more elements and solve large, consequential problems.”
Researchers have created networks with small light sources at each neuron that broadcast optical signals to thousands of connections, doing away with traditional electronic communication infrastructure, the release reported. This method can be particularly energy-efficient if photons, the smallest optical signal that could be utilized to indicate a spike, are detected using superconducting electronics.
In a recent work published in Nature Electronics, NIST researchers report creating a circuit that functions very similarly to a human synapse while using just single photons for signal transmission and reception. Superconducting single-photon detectors enable such a feat, according to the release. The Josephson junction, a superconducting circuit component, is where the computing takes place.
