Abstract
A new machine learning model is presented to predict the dynamic behavior of threshold voltage shifts induced by bias temperature instability (BTI) in CMOS devices. The model is constructed by combining physical theories with machine learning such as an artificial neural network and a Gaussian mixture model (GMM). To enlarge the capture–emission energy (CEE) window and to perform independent estimations of two distinct components of CEE distribution, the GMM with soft clustering is utilized, enabling full lifetime modeling of BTI. By training the CEE map with the consideration of the occupancy probability of traps and then executing the integration along the CEE, the threshold voltage shifts are obtained. This approach forms data-driven modeling that naturally encodes underlying physical theories as prior information. The resulting model exhibits a good performance for predicting the dynamic characteristics of BTI under various stress-recovery conditions.
Highlights
Accurately modeling bias temperature instability (BTI)-induced ΔVth relies on a combined methodology comprising thorough capture–emission energy (CEE) maps and statistical models
Accurately modeling BTI-induced ΔVth relies on a combined methodology comprising thorough CEE maps and statistical models
Purely data-driven Machine learning (ML) methods ignore any knowledge about the underlying physical processes and need to be coupled with physics-based modeling of BTI.37–39
Summary
ARTICLES YOU MAY BE INTERESTED IN Physics-informed machine learning for inorganic scintillator discovery The Journal of Chemical Physics 148, 241729 (2018); https://doi.org/10.1063/1.5025819 Impact of grain orientation and phase on Volta potential differences in an additively manufactured titanium alloy AIP Advances 11, 025219 (2021); https://doi.org/10.1063/5.0038114 Data-driven physics-informed constitutive metamodeling of complex fluids: A multifidelity neural network (MFNN) framework Journal of Rheology 65, 179 (2021); https://doi.org/10.1122/8.0000138
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have