User`s guide

3 Linear Model Identification
The prewhitening filter is determined by modeling the input as an
Auto-R egre ssive (AR) process of order N.Thealgorithmappliesafilter of
the form A(q)u(t)=u_F(t). That is, the input u(t) is subjec ted to an FIR
filter A to pr oduce the filtered signal u_F(t). Prewhitening the input by
applying a whitening filter before estimation might improve the q uality of
the estimated impu lse resp on s e g.
The order of the prew hitening filter, N,istheorderoftheA filter. N equals
the number of lags. The default value of N is
10, which you can also spe cify
as
[].
4 In the Model Name field, enter the name of the correlation analysis mo de l.
The name of the model should be unique in the Model Board.
5 Click Estima te to add this model to the Model B oard in the System
Identification Tool GUI.
6 In the Correlation Model dialog box, click Close.
7 To view the transient response plot, select the Transient resp check b ox
in the System Identification Tool GU I. For more information about working
with this plot and selecting to view im pulse- versus step-respo nse, see
“Using Impulse- and Step-Respon se Plots to Validate Models” on page 8-24.
You can export the model to the MATLAB w orkspace for further analysis
by dragging it to the To Workspace rectangle in the System Identification
Tool GUI.
How to Estimate Correlation Models at the Command
Line
You can use impulse and step commandstoestimatetheimpulseandstep
response directly from time- or frequency-domain data using correlation
analysis. Both
impulse and ste p produce the same FIR model, but generate
different plots.
Note cra is an alternative method for computing impulse response from
time-domain data only.
3-16