Wavecom W-SPECTRA (Automatic Monitoring System) V1.1.0 - Manual (Page 212)

202

Classifier – the Brain of W-SPECTRA WAVECOM Monitoring System W-SPECTRA, W-SPEED Manual V1.1.0
Insert or overwrite a database entry with classification results.
The actual range of classification depends on the frequency settings. Thus it is important that the mode
corresponds to the actual input signal frequency range.
The success of classification attempts depends mainly on the selection of the proper sampling time. To
recognize low speed signals, the higher sampling rate, i.e., 3.2 s should be selected.
It is worth keeping in mind that the classifier is based on statistically based estimates of the properties of
the input signals, and that many factors determine the reliability of these results – among these are signal
quality and application settings. However, together with common sense and good monitoring experience
and skills the classifier provides a powerful aid in analyzing unknown signals.
Classifier Parameters
There are following parameters in a classifier:
Modulation Check any number of modulation types or voice using the appropriate check boxes
to narrow down the range of recognized signals.
Continuous Mode If continuous classification is desired check this box. Continuous mode can be
halted and restarted pressing the Start/Stop Classifier button.
If the Continuous Mode check box is left unchecked, each classification must be
started manually by pressing the Start/Stop Classifier button. The classifier al-
lows only one classification attempt at a time. During classification attempts the
Start/Stop Classifier button is grayed out.
Refresh List If the Refresh List check box is checked, the list of results will be cleared after
each classification attempt. This feature is only available when continuous mode is
selected.
Cycle Mode This parameter sets the interval between classification attempts if no signals were
classified. Enter a restart cycle length ranging from 4 s to 3600 s. Samples are
available from two sources:
Samples taken before the classifier was started (select Previous Samples).