Wave-based device scaling concept for brain-like energy efficiency and integration

Abstract

Aiming at achieving brain-like capabilities and efficiencies with nano-scale integration, this paper proposes a new scaling concept for wave-based neuromorphic devices. Our scaling approach shows that vg the group velocity of a signaling wave is one of the essential parameters that can account for the orders-of-magnitude gaps in the efficiency of the human brain and VLSIs for cognitive tasks.