Development and Validation of an ANN-Based Approach for Temperature-Dependent Equivalent Circuit Modeling of SAW...

Area of Science:

• Electrical Engineering
• Materials Science
• Acoustics

Background:

• Surface Acoustic Wave (SAW) resonators are crucial components in RF systems.
• Accurate modeling of SAW resonator behavior across varying temperatures is essential for device performance.
• Traditional modeling approaches may not fully capture complex temperature dependencies.

Purpose of the Study:

• To develop a novel, temperature-dependent model for SAW resonators.
• To integrate artificial neural networks (ANNs) with lumped-element equivalent circuit models.
• To enable accurate RF characteristic simulations of SAW devices over a wide temperature range.

Main Methods:

• A lumped-element equivalent circuit model was combined with artificial neural networks (ANNs).
• ANNs were employed to model the temperature dependence of equivalent circuit parameters (ECPs).
• Scattering parameter measurements of a 423.22 MHz SAW device were used for validation from 0 °C to 100 °C.

Main Results:

• An ANN-based model accurately predicted the temperature-dependent behavior of the SAW resonator.
• The model successfully simulated RF characteristics within the tested temperature range (0 °C to 100 °C).
• The accuracy of the ANN-based model was found to be comparable to the original equivalent circuit model.

Conclusions:

• The proposed ANN-based approach provides an effective method for temperature-dependent SAW resonator modeling.
• This model eliminates the need for repeated measurements or equivalent circuit extraction for simulations.
• The developed model enhances the simulation capabilities for SAW devices under varying thermal conditions.