Wave Interference Functions for Neuromorphic Computing

Rahman, M. and Khasanvis, S. and Shi, J. J. and Moritz, C. A.. (2015) Wave Interference Functions for Neuromorphic Computing. Ieee Transactions on Nanotechnology, 14 (4). pp. 742-750.

Full text not available from this repository.


Neuromorphic computing mimicking the functionalities of mammalian brain holds the promise for cognitive capabilities enabling new intelligent applications. However, research efforts so far mainly focused on using analog and digital CMOS technologies to emulate neural activities, and are yet to achieve expected benefits. They suffer from limited scalability, density overhead, interconnection bottleneck and power consumption related constraints. In this paper, we present a transformative approach for neuromorphic computing with Wave Interference Functions (WIF). This is a framework using emerging nonequilibrium wave phenomenon such as spin waves. WIF leverages inherent wave attributes for multidimensional, multivalued data representation and communication, resulting in reduced connectivity requirements and efficient neural function implementations. It also yields a compact implementation of an artificial neuron. Moreover, since WIF computation and communication are in the spin domain, extremely low-power operation is possible. Our evaluations indicate upto 57xhigher density, 775xlower power and 2xbetter performance when compared to an equivalent 8-bit 45-nm CMOS neuron. Our scalability study using arithmetic circuits for higher bit-width neuron implementations indicate upto 63x density, 884x power and 3x performance benefits in comparison to a 32-bit CMOS equivalent design at 45 nm.

Item Type: Article
Additional Information: ISI Document Delivery No.: CM6SZTimes Cited: 0Cited Reference Count: 19Rahman, Mostafizur Khasanvis, Santosh Shi, Jiajun Moritz, Csaba AndrasDARPA Program on Nonvolatile Logic; Center for Hierarchical Manufacturing (CHM, NSF) at UMass AmherstThis work was supported by the DARPA Program on Nonvolatile Logic and the Center for Hierarchical Manufacturing (CHM, NSF) at UMass Amherst.0Ieee-inst electrical electronics engineers incPiscataway1941-0085
Uncontrolled Keywords: Physics
Collections: Nanomanufacturing Research Collection > Nanomanufacturing Nanoscale Science and Engineering Centers > Center for Hierarchical Manufacturing
Depositing User: Robert Stevens
Date Deposited: 12 Nov 2015 18:38
Last Modified: 12 Nov 2015 18:38
URI: http://eprints.internano.org/id/eprint/2328

Actions (login required)

View Item View Item