Low Power and Temperature-Resilient Compute-In-Memory Based on Subthreshold-FeFET

TL;DR

This paper introduces a temperature-resilient, ultra-low power compute-in-memory design using subthreshold FeFETs that maintains accuracy across a wide temperature range and significantly improves energy efficiency for AI applications.

Contribution

It proposes a novel 2T-1FeFET CiM architecture that is robust to temperature variations and operates efficiently at subthreshold voltages, advancing low-power AI hardware.

Findings

Achieves 89.45% CIFAR-10 accuracy with VGG model.

Demonstrates immunity to temperature drift at 8-bit scale.

Attains 2866 TOPS/W energy efficiency.

Abstract

Compute-in-memory (CiM) is a promising solution for addressing the challenges of artificial intelligence (AI) and the Internet of Things (IoT) hardware such as 'memory wall' issue. Specifically, CiM employing nonvolatile memory (NVM) devices in a crossbar structure can efficiently accelerate multiply-accumulation (MAC) computation, a crucial operator in neural networks among various AI models. Low power CiM designs are thus highly desired for further energy efficiency optimization on AI models. Ferroelectric FET (FeFET), an emerging device, is attractive for building ultra-low power CiM array due to CMOS compatibility, high ION/IOFF ratio, etc. Recent studies have explored FeFET based CiM designs that achieve low power consumption. Nevertheless, subthreshold-operated FeFETs, where the operating voltages are scaled down to the subthreshold region to reduce array power consumption, are particularly vulnerable to temperature drift, leading to accuracy degradation. To address this challenge, we propose a temperature-resilient 2T-1FeFET CiM design that performs MAC operations reliably at subthreahold region from 0 to 85 Celsius, while consuming ultra-low power. Benchmarked against the VGG neural network architecture running the CIFAR-10 dataset, the proposed 2T-1FeFET CiM design achieves 89.45% CIFAR-10 test accuracy. Compared to previous FeFET based CiM designs, it exhibits immunity to temperature drift at an 8-bit wordlength scale, and achieves better energy efficiency with 2866 TOPS/W.