Different results between psd_welch in labview and pwelch in matlab

[ttrr]

Hi,

I use the following code. Both MathScript and TSA return 129 data of PSD.

For the Simulate Signal express VI, the settings are

Frequency: 10.1 Hz

Amplitude: 1

Phase: 0

Offset: 0

Sampling Rage: 10000 Hz

Number of Samples: 1000 (Auo)

[pem]

Thank ttrr, I used waveform chart to show the psd and gave wrong value

[Wenxu]

I came across the same problem and still cannot figure out why the pwelch function gives a different result with labview TSA welch function. Interestingly, I found the FFT psd also has a very difference with TSA welch function. I think the two psd calculation methods though differ from each other but should have results similar with each other. Ohterwise, which function should we trust? 

Regards

Lwx

[sujithsb]

Hi,

 I am doing project on labview.

and i am working on signal processing welch window.

so can you please send me the vi what you have done, it may help me a lot.

[GerdW]

Hi sujith,

you have been asking this question several times before. [1] [2] [3]

You have been told not to create duplicate post for the very same topic - that includes not to hijack older threads...

You were asked questions which you haven't answered until now...

Why don't you just recreate the snippets shown in this thread?

[MengHuiHanTang]

psd_welch is not the only function which returns different results from those of Matlab. Just lately I had the same trouble using 'contourc' function. MathScript 'contourc' gives a slightly different contour than the one from Matlab. I was told they were both right, because the results were so close.

Here is the link to that post http://forums.ni.com/t5/LabVIEW/Results-from-contourc-in-MathScript-not-match/td-p/2436236

Combined with this post, it seems that MathScript implementation of m code has more issues than I previously thought. 

[Zach-H]

@ MengHuiHanTang: If you re-read the last two posts by ttr and pem you'll note that both the MathScript node and TSA did return 129 data elements as well. The difference was due to using a waveform chart rather than a graph.