PROJECT TITLE :

A CMOS Spiking Neuron for Brain-Inspired Neural Networks With Resistive Synapses and In Situ Learning

ABSTRACT:

Nanoscale resistive memory devices are expected to fuel dense integration of electronic synapses for large-scale neuromorphic systems. To realize such a brain-galvanized computing chip, a compact CMOS spiking neuron that performs in situ learning and computing whereas driving a large variety of resistive synapses is desired. This temporary presents a novel leaky integrate-and-fire neuron style that implements the dual-mode operation of current integration and synaptic drive, with a single operational amplifier (opamp) and permits in situ learning with crossbar resistive synapses. The proposed style was implemented in a very 0.18- CMOS technology. Measurements show neuron's ability to drive a thousand resistive synapses and demonstrate in situ associative learning. The neuron circuit occupies a little area of zero.01 mm a pair of and has an energy efficiency worth of 9.three pJ/spike/synapse.