Effects of Temperature and Device-to-Device Variability in pFET-Based Bias Temperature Instability Reservoir Computing

Abstract

We assess the reliability and sensitivity to device-to-device variability of our pFET-based Bias Temperature Instability (BTI) Physical Reservoir Computing system. We examine the device's response at different temperatures (25 °C and 75 °C), noting a gradual current drop when the device operates at higher temperatures. Furthermore, we perform gait authentication tests across various sizes, temperatures, and device sets to evaluate the performance of our system. Additionally, we utilize a stochastic model based on the Compact-Physical (Comphy) BTI physical model to further analyze our system. This model not only clarifies the underlying physics but also assists in designing and selecting the optimal reservoir for a range of applications.