User`s guide

3 Linear Model Identification
You can choose from the following three spectral-analysis methods:
•
etfe (Empirical Transfer Function Estimate)
For input-output data. This method computes the ratio of the Fourier
transform o f the output to the Fourier transform of the input.
For time-series data. This method computes a periodogram as the
normalized absolute squares of the Fou rier transform of the time series.
ETFE works well for highly resonant systems or narrowband systems.
The drawback of this method is that it requires linearly spaced f requency
values, d oes not estimate the disturbance spectrum, and does not provide
confidence intervals. ETFE also works well for periodic inputs an d
computes ex act estimates at multiples of the fundamental freque ncy of the
input and their ratio.
•
spa (SPectral Analysis)
This method is the B lackman-Tukey spectral a nalysis m ethod, w h ere
windowed v ersions of the covariance functions are Fourier transformed.
For more information about this algorithm , see “Spectral Analysis
Algorithm” on page 3-8.
•
spafdr (SPectral Analysis with Frequency Dependent Resolution)
This method is a variant of the Blackman-Tukey spectral analysis method
with frequency-dependent resolution. First, the algorithm computes
Fourier transform s of the inputs an d outputs. N ext, the products of the
transformed inputs and outputs with the conjugate input transform are
smoothed over local frequency regions. The widths of the local frequency
regions can vary as a function of frequency. The ratio of these averages
computes the frequency-response estimate.
Options for Frequency Resolution
• “What Is Frequency Resolution?” on page 3-7
• “Frequency Resolution for etfe and spa” on page 3-7
• “Frequency R esolution for spafdr” on page 3-7
• “etfe Frequency Resolution for Periodic Input” on page 3-8
3-6