Design of cellular neural network architecture using memristors

Abstract

A method to design a Cellular Neural Network architecture using memristors is proposed. Memsitor is a new resistive device whose resistivity varies depending upon applied voltage/current. In this paper, we show that a memristor bridge circuit consisting of four identical memristors that is able to perform zero, negative and positive synaptic weightings. Together with three additional transistors, the memristor bridge weighting circuit is able to perform synaptic operation for neural cells. Also, we show that Cellular Neural Networks can be implemented with the programming of different set of weights on the memristor bridge synapses.