User`s guide
7 Recursive Techniques for Model Identification
Data Segmentation
For systems that exhibit abrupt changes while the data is being collected,
you might want to develop models for separate data segm ents such that the
system does not change during a particular data segment. Such modeling
requires identification of the time instants w hen the changes occur in the
system, breaking up the data into segments according to these time instants,
and identification of mo de ls for the diffe rent data segments.
The following cases are typical applications for data segmentation:
• Segmentation of speech signals, where each data segment corresponds
to a phonem.
• Detection of trend breaks in tim e series.
• Failure detection, where the data segments correspond to operation with
and without failure.
• Estimating different working modes o f a system.
Use
segment to build polynomial models, such as ARX, ARMA X, AR, and
ARMA, so that the mo de l parameters are piece-wi se cons tant over tim e. For
detailed information about this comm and, see the corresponding reference
page.
To see an example of usin g data segmentation, run the Recursive Estim ati on
and Data Segme ntatio n demonstration by typing to the following command at
the prompt:
iddemo5
7-14