Optimizations for a Current-Controlled Memristor-based Neuromorphic Synapse Design

TL;DR

This paper presents an optimized current-controlled memristive synapse circuit that enhances energy efficiency, reliability, and performance in neuromorphic systems, demonstrated through a 4-bit SNN application with significant improvements over prior designs.

Contribution

The paper introduces a novel current-controlled memristive synapse design that improves energy efficiency and reliability, especially under process variations and stochastic memristor behavior.

Findings

Up to 82% energy savings during SET operations.

Up to 54% energy savings during READ operations.

2.7x improvement in classification accuracy in neuromorphic applications.

Abstract

The synapse is a key element of neuromorphic computing in terms of efficiency and accuracy. In this paper, an optimized current-controlled memristive synapse circuit is proposed. Our proposed synapse demonstrates reliability in the face of process variation and the inherent stochastic behavior of memristors. Up to an 82% energy optimization can be seen during the SET operation over prior work. In addition, the READ process shows up to 54% energy savings. Our current-controlled approach also provides more reliable programming over traditional programming methods. This design is demonstrated with a 4-bit memory precision configuration. Using a spiking neural network (SNN), a neuromorphic application analysis was performed with this precision configuration. Our optimized design showed up to 82% improvement in control applications and a 2.7x improvement in classification applications compared with other design cases.