how can I extract a gaussian function out of a spectrum which has five gaussain functions....

[altenbach]

Just for completeness, heres how you would fit the data using the general linear filt, which assumes that only the amplitudes are unknown.

(There was a formula error in my previous version, whcih I have corrected above).

Also here, you can add more gaussians by simply adding rows to the array of clusters. As you can see, the fit is not great.

I also show the component spectra.

[Raulgoz10]

That is perfectly working now. Thank you very much. but is it possible to plot `data, best fit, the final gauss functions and original spectrum minus final gauss functions`` rather than ``data, best fit and guess´` ?

[altenbach]

---

@zola1024 wrote:

 but is it possible to plot `data, best fit, the final gauss functions and original spectrum minus final gauss functions`` rather than ``data, best fit and guess´` ?

---

Yes, everything is possible.

[altenbach]

Try this:

 

[altenbach]

Just on a sidenote, notice that my gaussians are normalized for unit height, independent of width. It would typically be more reasonable to normalize for unit area. It would be trivial to adjust the model function accordingly.

[grimph1]

Hello,

would it be possible to convert your multiple gaussian fit vi to labview version 9.0.1.

Unfortunately i'm not able to open it.

Thanks in advance.

[altenbach]

---

grimph1 wrote:

would it be possible to convert your multiple gaussian fit vi to labview version 9.0.1.

---

Try this...

[grimph1]

Thank you very much.

Unfortunately i receive an error code 3 concerning a problem while frontpanel is loaded.

Can you help ?

[altenbach]

Sorry, I have no way of testing the downconverted VI in LV2009. Also error 3 seems very odd. Did you extract the zip file first?

What is the full error message?

Here's another save, this time with slightly different options. See if this works better....