Frequency-Domain Models for Nonlinear Microwave Devices Based on Large-Signal Measurements

Area of Science:

• Electrical Engineering
• Nonlinear Circuit Analysis
• Electromagnetics

Background:

• Traditional scattering (S)-parameters are limited to linear systems.
• Analyzing nonlinear circuits requires advanced characterization techniques.
• Existing nonlinear models can be complex to derive or unavailable.

Purpose of the Study:

• Introduce nonlinear large-signal scattering (S)-parameters as a novel frequency-domain mapping.
• Develop a general framework for nonlinear large-signal S-parameters, impedance (Z), and admittance (Y) parameters.
• Provide a practical tool for nonlinear circuit design and analysis.

Main Methods:

• Formulation of general nonlinear large-signal S-, Z-, and Y-parameters.
• Derivation of inter-conversion equations between parameter sets.
• Application in the design of a 1 GHz frequency-doubler circuit.
• Development of an extraction method using artificial neural networks and nonlinear vector network analyzer measurements.

Main Results:

• Nonlinear large-signal S-parameters generalize classic S-parameters for nonlinear systems.
• Demonstrated utility in designing a specific nonlinear frequency-doubler.
• Successful extraction of parameters using ANNs when nonlinear models are unavailable.
• Nonlinear large-signal S-parameters found to be more general than nonlinear scattering functions.

Conclusions:

• Nonlinear large-signal S-parameters provide a powerful and versatile tool for characterizing nonlinear circuits.
• The proposed extraction method enables parameter determination even without pre-existing nonlinear models.
• This work advances the analysis and design capabilities for complex nonlinear electronic systems.