Neural network model speeds ferroelectric transistor circuit simulations

NN-Based FeFET Compact Modeling for SPICE Simulation

Emerging Technologies

Summary

Ferroelectric transistors are new devices that need fast and accurate models for testing electronic circuits. Traditional models require complex physics and take a long time to build. The authors use neural networks to create a model that learns from actual measurements, even when the data is noisy. Their model works well in common circuit simulators to predict how memory devices behave in large arrays. This method can make it quicker and easier to design circuits using emerging transistor technologies.

What this means in practice

  • For circuit designers: Use the neural network model to simulate large ferroelectric memory arrays in SPICE for early prototype evaluation and design optimization.
  • For device model developers: Integrate measurement-based neural network models to reduce development time and handle noisy experimental data in transistor compact modeling.

Authors

Rodion Novkin, Simon Thomann, Hussam Amrouch

Abstract

Prototyping and early-stage circuit evaluation for emerging transistor technologies require fast and accurate compact models for SPICE simulations. Standard physics-based compact models, such as BSIM, offer high fidelity but demand extensive domain expertise and time-consuming equation formulation for every new physical effect, significantly slowing technology development. Neural network (NN)-based models offer a scalable alternative approach, but they rely heavily on smooth numerical solver data with explicit high-order derivative information. However, measurement data suffers from complex noise patterns, making derivative extraction challenging and complicating NN-based transistor modeling. This work presents a derivative-free NN training framework employing spectral normalization to naturally smooth derivative behavior and filter measurement noise. Using drain current measurements from a fabricated 28 nm ferroelectric device, we design an NN-based compact model that captures the full spectrum of polarization amplitudes and supports dynamic polarization switching in transient simulations. To demonstrate its capability, we deploy the model in a commercial SPICE software to simulate write and read operations for large-scale memory arrays with up to 512x512 size, including parasitic effects. Our approach bridges raw measurements with NN-based SPICE-compatible transistor modeling, significantly reducing model development time and enabling rapid circuit-level prototyping for emerging devices.