Analytical Modelling of Y2O3-based Memristive System for Artificial Synapses

Abstract

Artificial synapses are the key units for information processing in neuromorphic systems. Memristive systems are frequently used as an artificial synapse because of their simple structures, gradually changing conductance and high-density integration. In this work, a non-linear analytical model for Y2O3-based memristive system with new parabolic window function has been discussed for artificial synapses applications. Moreover, resistive switching characteristic and synaptic plasticity properties of the memristive systems are modelled by utilizing non-linear analytical model to investigate the performance of artificial synapse. Further, the modelled data is verified by the experimental results of fabricated devices which confirmed that the developed model can be realized the basic functions of spiking neurons and has great potential for neuromorphic computing. © 2020 IEEE.