Spike-Based Hybrid Computers

Abstract

This report documents progress made during the period July 1st 2002 to Jan 2003 for the ONR grant 'Spike-Based Hybrid Computers'. It is the fourth and final progress report for this grant. Our progress this year was in three areas: (1) Ultra-Low-Power Time-Based Analog-to-Digital Conversion inspired by neuronal signal representations; (2) An analog integrated circuit memory element for use in learning and adaptive systems, and (3) A predictive comparator with adaptive control that exploits knowledge of the signal to predict it and thus save energy.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Oct 02, 2003
Accession Number
ADA417467

Entities

People

  • Rahul Sarpeshkar

Organizations

  • Massachusetts Institute of Technology

Tags

Communities of Interest

  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Adaptive Systems
  • Analog Signals
  • Comparators
  • Computations
  • Computers
  • Difference Equations
  • Electronics
  • Energy
  • Energy Consumption
  • Equations
  • Hybrid Computers
  • Integrated Circuits
  • Learning
  • Military Research
  • Power Electronics
  • Power Supplies
  • Signal Processing

Readers

  • Integrated Circuit Design and Technology.
  • Neural Network Machine Learning.
  • Technical Research and Report Writing.