存在温度、工艺变化、器件老化和闪烁噪声的铁电体FinFETs类脑计算
Neuromorphic Computing with Ferroelectric FinFETs in the Presence of Temperature, Process Variation, Device Aging and Flicker Noise
摘要：
针对温度变化、工艺变化、闪烁噪声和器件老化对预训练全铁电FinFET深神经网络推理精度的影响进行了全面研究。在厚度为5 nm的铪锆氧化物（HZO）基场效应FinFET器件中，采用一种新颖的自适应编程读算法实现了100 ns写脉冲下的多电平单元（MLC）操作。以MNIST数据集上具有97.5%推断准确度的预训练神经网络（NN）作为基线，器械间差异的影响可忽略不计。在各种偏置条件下的闪烁噪声特性表明，漏极电流波动小于0.7%，几乎没有推理精度下降。一个程序化的细胞，作为一个后果的温度变化，电导漂移被捕获的一个紧凑的模型在很宽的范围内的栅极偏置。尽管在300 K下训练的NN在233 K下的推理精度显著下降，但证明了用于恢复精度的门偏置优化。显示了超过10个$^8$循环的耐久性和超过10年的外推保持率，这为使用FE-FinFET的边缘器件人工智能铺平了道路。
英文摘要：
This paper reports a comprehensive study on the impacts of temperature-change, process variation, flicker noise and device aging on the inference accuracy of pre-trained all-ferroelectric (FE) FinFET deep neural networks. Multiple-level-cell (MLC) operation with a novel adaptive-program-and-read algorithm with 100ns write pulse has been experimentally demonstrated in 5 nm thick hafnium zirconium oxide (HZO)-based FE-FinFET. With pre-trained neural network (NN) with 97.5% inference accuracy on MNIST dataset as baseline, device to device variation is shown to have negligible impact. Flicker noise characterization at various bias conditions depicts that drain current fluctuation is less than 0.7% with virtually no inference accuracy degradation. The conductance drift of a programmed cell, as an aftermath of temperature change, was captured by a compact model over a wide range of gate biases. Despite significant inference accuracy degradation at 233K for a NN trained at 300K, gate bias optimization for recovering the accuracy is demonstrated. Endurance above 10$^8$ cycles and extrapolated retention above 10 years are shown, which paves the way for edge device artificial intelligence with FE-FinFETs.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
