User`s guide

Identifying State-Space Models
Data Supported b
y State-Space Models
• “Types of Suppor
ted Data” on page 3-77
• “Estimating Con
tinuous-Time Models” on page 3-77
• “Designating Da
ta for Estim a t in g Di screte-Time Models ” on page 3-78
Ty pes of Supported Data
You can estimate
linear state-space models from data with the following
characteristic
s:
• Real data or com
plex data in any domain
• Single-output a
nd multiple-output
• Time- or frequen
cy-domain data
To estimate stat
e-space m odels for time-series data, see Chapter 6, “Time
Series Model Id
entification”.
You must first im
port your data into the MATLAB workspace, as described in
Chapter 1, “Dat
a Processing”.
Estimating Continuous-Time Models
Use either of t
he following ways to e stimate continuous-time, state-space
models:
• To get a linear
, continuous-time model of arbitrary structure for
time-domain d
ata, you can estimate a d iscrete-time model, and then use
d2c to transfo
rm it to a continuous-time model.
• Use continuo
us-time frequency-domain data.
To denote cont
inuous-time frequency-domain data, set the data sampling
interval to 0
. You can set the sampling interval when you import data into
the GUI or set
the
Ts property of the data object a t the command line .
3-77