DESIGN OF AN ADAPTABLE NEUROMORPHIC System for Emerging NanOELECTRONIC Options (ANSENO)

Abstract

The objective of the proposed fundamental research is to demonstrate a hybrid CMOS-nanoelectronic system for evaluating novel nanoscale devices in the context of neuromorphic algorithms. The system is adaptable in the sense that various nanoelectronic devices can be integrated into upper metalization layers for the purpose of experimentally evaluating their utility and performance for neuromorphic applications. The system is reconfigurable, allowing for the implementation of any neuromorphic application in the form of a spiking neural network that can be mapped to the resulting hardware. This work will be accomplished across three layers of abstraction: (1) device/circuit co-design of CMOS circuits ready for integration with a wide range of novel nanoelectronic device technologies; (2) architectural design and system integration of the neuromorphic computing and testing system; and (3) development of a software development kit, with emulation models and tools used for training neural networks. This design-focused proposal is submitted in conjunction with a parallel proposal from the SUNY Polytechnic Institute, that will focus on fabrication efforts, led by Prof. Nathaniel Cady. This proposal fits within the larger AFRL-RI component of an Applied Research for the Advancement of Science and Technology Priorities (ARAP) effort.

Document Details

Document Type
DoD Grant Award
Publication Date
Oct 01, 2021
Source ID
FA87502111018

Entities

People

  • Garrett Steven Rose

Organizations

  • Rome Laboratory
  • United States Air Force
  • University of Tennessee

Tags

Readers

  • Integrated Circuit Design and Technology.
  • Neural Network Machine Learning.
  • Research Science/Academic Research

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks
  • Biotechnology