Simulation of coupled spin torque oscillators for pattern recognition

Abstract

In this paper, we use circuit-level simulations to investigate the synchronization dynamics of spin torque oscillators (STOs) and demonstrate a pattern recognition scheme based on STO dynamics. We perform a sensitivity analysis in order to determine the robustness of the different STO coupling methods, considering parameter variations, such as radius or thickness of STOs. After pointing out the advantages of the cross coupled architecture, we demonstrate a coupling scheme for pattern recognition. Several patterns are encoded in the resistive coupling network and are successfully identified, after the network has been initialized correspondingly. Finally, by extracting the corresponding impulse sensitivity function, we show how an existing analytical phase domain model can be adapted to the STO array. We compare the results of this analytical model with the numerical model of the STO and conclude that they agree well providing thus a fast method for design.

Document Details

Document Type
Pub Defense Publication
Publication Date
Oct 10, 2018
Source ID
10.1063/1.5042423

Entities

People

  • Amir Hossein Fallahpour
  • Bogdan Popescu
  • Dan Popescu
  • Gyorgy Csaba
  • Markus Becherer
  • Paolo Lugli
  • Wolfgang Porod

Organizations

  • Centro Universitario da FEI
  • Defense Advanced Research Projects Agency
  • German Research Foundation
  • Pázmány Péter Catholic University
  • Technical University of Munich
  • University of Notre Dame

Tags

Fields of Study

  • Physics

Readers

  • Computational Modeling and Simulation
  • Computer Vision.
  • Integrated Circuit Design and Technology.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Neural Networks