reverse PSD

[Aris1983]

Hi all 

I am now in a position wanting to reverse PSD. I have used the following vi and the values I have out of and accelerometer are PSD values and I want to get back to the original values that came out of the accelerometer (g). Is there any way of doing this?

Thanks

[Questionmarker]

Hey,

although I do not have all the VIs you use in your VI from what I see I would judge the following:

- you seem to get Raw Data and use the "Spectral Measurements.vi" to get a PSD. Furthermore, you also use the "SVFA PSD".

Both use raw Data. If you want to reverse the PSD, why not just take the raw data?

From the Math of a PSD I have to tell you, there is no reverse, as you killed the phase information during the process. Also, you might have applied a filter like "Hanning" in your example. If you want to reverse, use Amplitude and Phase, or x,y or whatever representation supplied. Calculating back and forth will introduce some edge errors (Gibbs Phenomenon in english I guess).

So, I think the question you want to ask might be a bit different - give reason what you want to calculate backwards and why - maybe just applying a filter might suffice?

Regards!

[Aris1983]

In the stored data there s no sign of the raw data....seems that something went wrong in storing...so now i want to get the raw data back...

[Joel_G]

Hi Aris,

If you do the fast fourier transfrom to get the Power Spectrum, can't you just do the Inverse fourier transform to get it back to the signal? There is a function for this called inverse fourier transform.vi.

Kind Regards,

[Questionmarker]

Hey Joel and Aris,

if you do a Fourier Transformation you can reverse it. If you do an FFT, you can (more or less) reverse it. Sadly, the PSD involves a quadrature of the FFT (which usually yields a complex result) - please see e.g. http://en.wikipedia.org/wiki/Spectral_density for details, so there is no possible way back to the raw data. So if you want raw data and do not care for filters and so on, i guess this is where your raw data are at:

oh while painting the picture I realized, you already have your raw data - they are in the last signal channel. I guess you solved your problem by yourself - sometimes the "clean up diagram button" actually helps (;

Kind Regards!