'Neuromorphic Computing' could lead to more energy-efficient computing

These 'neuristor' circuits, created using niobium dioxide (NbO2), replicates the switching behavior observed in ion channels within biological neurons in the human brain, which can perform complex computations using an incredibly small amount of power



Researchers from the department of materials science and engineering at Binghamton University, want to create more energy-efficient computers, so devices like drones could be receptive to their surrounding atmosphere, without being disturbed about a Wi-Fi signal linking them to a larger computer machine.


These NbO2 devices are produced by applying a large voltage across a non-conductive niobium pentoxide (Nb2O5) film, initiating the formation of conductive NbO2 filaments which are responsible for the important switching behavior. But this high-voltage and time-consuming post-fabrication process makes it difficult to produce dense circuits, which are needed for complex computer processors.


©2018 by Technoscious