User`s guide

Identifying Hammerstein-Wiener Models

If only the input nonlinearity is present, the m odel is called a Hammerstein

model. If only th e outpu t non li nearity is present , the model is c al led a Wiener

model.

You must specify the follow ing model orders for the linear block:

• n

—The number of zeros plus one.

• n

—The number of poles.

• n

—The delay from input to the output in terms of the number of samples.

For ny outputs and nu inputs, n

, n

,andn

are ny-by-nu matrices. You can

specify a nonlinearity for only certain inputs and outputs, and exclude the

nonlinearity for other inputs and outputs.

Nonlinearity Estimators for Hamm erstein-Wiener

Models

Hammerstein-Wiener models support the following nonlinearity estimators

for estimating the parameters of its input and output nonlinear blocks:

• Dead Zone

• Piecewise Linear

• Saturation

• Sigmoid N etwork

• Wavelet Network

• One-Dimensional Polynomial

• Unit Gain

• Custom Network

For a summary of all nonlinearity estimators and links to the corresponding

reference pages, see “Supported Nonlinearity Estimators” on page 4-25.

You can exclude either the input nonlinearity or the output nonlinearity from

the model structure:

4-17