SigThresh

Author : Ian Taylor, Bernard Schutz

Version : 2.0

Input Types : GraphType

Output Types : GraphType

Date : 16 March 2001


 

Contents


Description of SigThresh

The SigThresh unit allows the user to test data sets by comparing a quantity derived from their data to a given threshold value. If the data set passes the test then the input data is passed to the units second output node otherwise it is passed to the first. You can use this unit to process the data using different networks if a certain condition arises. For example, you may be waiting for an event to appear in the data and then when this event occurs you may want to process it and graph it in a particular way.

There are 5 different comparison tests you can use on the input data sets. You can choose to use the mean, standard deviation, or RMS of the entire data set, or you can compare the value or the absolute value of each data point against the threshold value. In the value test, if the given threshold value is negative, then the data are tested to see if they exceed it in the negative sense, i.e. are smaller than the threshold.

Input data sets can be any GraphType objects. That means they can contain multidimensional data arrays. All elements of such arrays are used in constructing the tests.

If the data set is complex, its real part is used for the value and mean tests, while the complex magnitude is used for the absolute value, standard deviation and RMS tests.

If the data set contains more than one dependent data set, only the first arithmetic data set is used.
 
 

Using SigThresh

The parameter window of SigThresh looks like this:  


 
The first line is a choice box offering the five different threshold tests described above. The second line allows you to set the value of the threshold.

If you want the threshold to depend on values calculated elsewhere, you can use the input node window of SigThresh to create an input node for the threshold value.