User`s guide

7 Recursive Techniques for Model Identification
What Is Recursive Estimation?
Many real-world applications, such as adaptive control, adaptive filtering, and
adaptive prediction, require a model of thesystemtobeavailableonlinewhile
the system is in operation. Estimating models for batches of input-output
data is useful for addressing the following types of questions regarding
system operatio n:
• Which input should be applied at the next sam pling instant?
• How should the parameters of a matched filter be tuned?
• What are the predictions of the next few outputs?
• Has a failure occurred? If so, what type of failure?
Youmightalsouseonlinemodelstoinvestigatetimevariationsinsystem
and signal properties.
The methods for computing online mode ls are called recursive identification
methods. R ecursive algorithms are also called recursive parameter estimation,
adaptive param eter estimation, sequential estimation,andonline algorithms.
For examples of recursive estimation and data segmentation, run the
Recursive Estimation and Data Segmentation demo by typing the following
command at the prompt:
iddemo5
For detailed inform ation abo u t recursi ve parameter estimation a lgorithms,
see the corresponding chapter in System Identification: Theory for the User by
Lennart Ljung (Prentice Hall PTR, Upper Saddle River, NJ, 1999).
7-2