3. Parametric identification

The System Identification toolbox in Matlab ^® offers a set of powerful tools dedicated to parametric identification. The general aim of parametric identification is to synthesize a model from measured input/output data. The starting point for identification is therefore the collection of a set of data consisting of commands sent to the process and measured output values at the same points in time (discrete-time identification). The control signal must be well chosen to excite the modes and operating points we wish to model. A pseudo-random binary sequence is commonly used for the frequency richness it provides.

In this section, it is assumed that the data have been collected with any measurement system and are available in Matlab ^® . Note that the dSPACE ^...