I had a similar problem in that I had spreadsheet data (x=timestamp in ms, y=voltage)which was sampled at a non-uniform sample rate (non-deterministic sampling loop), stored in a spreadsheet, that I wanted to operate upon using the waveform tools in labview, but the waveform data type requires a constant dt.
I solved this problem by first calculating the average sample rate of the original data, then interpolated the y values onto these new timestamps (now of equal dt).
The VI I used to do this is below, as well as the sample dataset I used to test the vi. My application needed double precision numbers reading from the spreadsheet, hence the re-written read from spreadsheet file vi included.
The sample dataset is simply two columns representing time (ms) and voltage. I simulated a sinusoidal voltage. I randomised the sample time so that the dt between samples is 6ms +-(some fraction of 2.5ms at random) - ie a non-uniform dt (the sample file contains extra columns used to produce the simulated dataset).
The VI creates a waveform of the interpolated data. It also plots the original data in x-y format alongside the interpolated data in x-y format so you can see the interpolation errors and non-uniform vs uniform dt's of the two datasets.
unzip the zip file, Run the VI "resample example.vi" (labview ver 7.1), browse to the text file in the zip which contains the test data.
look at the block diagram for notes.
Ive probably done overkill on this - no doubt someone will come up with a more elegant way of making non-uniform-dt data into a waveform type (or just tell me to find a way to sample at uniform dt in the first place!)
Steve
