Devices used in communication systems, such as traveling-wave tubes (TWTs) and solid-state power amplifiers (SSPAs), are usually described by analytical models that typically involve solving a set of simultaneous, nonlinear, partial differential equations by numerical methods. Unless some suitable simplifications can be made, such detailed models are too complex and computationally demanding to be useful in a system level simulation, where the nonlinear device is just one of many subsystems.
A higher-level model that converts an input waveform to (nearly) the correct output waveform, without necessarily resorting to the fundamental physics of the device, is needed. For linear systems, the transfer function is such a model. For nonlinear systems, the nonlinearity is represented either as a functional relationship or in tabular form for simulation applications. This representation is referred to as a behavioral model. It is a black-box approach to system level modeling, which provides a convenient mean of predicting system level performance without the computational complexity of full circuit model simulations.
Behavioral models can generally be divided into three groups: memoryless models, quasi-memoryless models, and behavioral models with memory. This chapter discusses how to model memoryless and quasi-memoryless nonlinear systems.