NB: this is taken from the IBM NEWDSP documentation. The timebase parameters in Jetdsp are still interpreted in this way (for IBM expressions) and are available by clicking on the 'Interpolation' button.

Processed signals always exist on a common "time" vector.

The raw data making up the signal is interpolated onto a common time vector with the number of points used dependent on the Datapoints: Number field. Either over the common timebase range or the timebase range as specified in the From and To fields, if the value entered for the number of points is zero (or >8192) then the superset of time points in all input traces is used up to a maximum of 8192. If the number of points exceeds this the value 8192 points are used, spaced according to the Datapoints: Spacing field. If the requested number of datapoints > 0 and < 8192 then the input value will will be used, again spaced according to the Datapoints: Spacing field.

Where the superset of all input datapoints is not used, timepoints are selected by one of three spacing methods:

Quartile:default spacing according to percentage of points in each quartile of the input range. Therefore reflects timepoint clustering in in each quartile.
Equal: equally spaced points throughout the range.
Gauss: randomly spaced points based on a gaussian distribution centred on the midpoint of the range.
The actual interpolant method to be used may be set as either CUBIC or LINEAR (default) in the Interpolant: Method+field.

IBM style signal processing expressions home page

Top JETDSP Home Page Manual FAQ
Please send comments to: data-and-coding-support@ccfe.ac.uk