Fourier transform -> power spectrum

AI Thread Summary
The discussion focuses on the relationship between the discrete Fourier transform (DFT) and the power spectrum of a signal. The DFT represents how a signal's frequency components are distributed, while the power spectral density (PSD) is derived from the square of the DFT's magnitude, indicating how power varies with frequency. It explains that periodic signals can be expressed as a sum of harmonics, with the fundamental frequency having the highest intensity. The inverse Fourier transform can be viewed as a sum of closely spaced harmonics, linking the amplitude of these harmonics to the energy spectral density. Overall, the conversation clarifies the connection between Fourier transforms and the power spectrum, enhancing understanding of signal analysis.
juef
Messages
27
Reaction score
0
Fourier transform --> power spectrum

Hey all!

I've been learning about the discrete Fourier transform (and FFT too) recently. What I don't understand is why applying it to a signal gives its power spectrum. I am not really good in physics, so to me it just seems like a magical formulae, one might say. :)

I don't need a deep explanation; the basics would be more than enough to please me. It would already be much more than what I can find on math sites and books about the discrete Fourier transform.

Thank you!
 
Physics news on Phys.org
The inverse Fourier transform represents a signal as a sum (continuous or discrete) of frequency, each of one having an amplitude.

The (direct) Fourier transform represents this repartition of frequency from the signal. It describes how a signal is distributed along frequency.

The power spectral density (PSD) (or spectral power distribution (SPD) of the signal) are in fact the square of the FFT (magnitude). It describes how the power of a signal is distributed with frequency.
 
A periodical signal can be written as a sum of harmonics (using the harmonic form of the Fourier series), each harmonic having it's own intensity (which decreases as the frequency of the harmonic increases). This harmonics are not abstract mathematical stuff. You can observe them while trying to set a tv channel and you see that you can catch that channel on several frequencies (these are the harmonics). The best quality will have the first harmonic (the fundamental) because it has the highest intensity. If you set the channel on the 2nd harmonic you will see that it will be noisy, and if you set it on the 3rd it will be even more noisy (because it's intensity is lower).
Now if you think of the inverse Fourier transform (which is an integral) as of a sum of harmonics of very close frequencies you can find an analogy: the direct Fourier transform of the signal (found in the formula of the inverse transform) plays the same role as the amplitude of the harmonics in the Fourier series. That's why it is called "spectral density of the complex amplitude". As you know, the energy is proportional to the square of the amplitude(intensity) so, the energy spectral density will be the square of the modulus of the Fourier transform (the Fourier transform has complex value). If you consider that the signal is applied on an ideal resistor of 1 ohm you can say that the square of the modulus of the Fourier transform is the spectral density of the power.
 
Thank you both for this enlightment! I believe I understand much better now.

antonantal said:
As you know, the energy is proportional to the square of the amplitude(intensity)[...]

Ah, well, no I didn't :shy: I should have mentioned that I'm studying maths and not physics. Thanks again!
 
Hi there, im studying nanoscience at the university in Basel. Today I looked at the topic of intertial and non-inertial reference frames and the existence of fictitious forces. I understand that you call forces real in physics if they appear in interplay. Meaning that a force is real when there is the "actio" partner to the "reactio" partner. If this condition is not satisfied the force is not real. I also understand that if you specifically look at non-inertial reference frames you can...
I have recently been really interested in the derivation of Hamiltons Principle. On my research I found that with the term ##m \cdot \frac{d}{dt} (\frac{dr}{dt} \cdot \delta r) = 0## (1) one may derivate ##\delta \int (T - V) dt = 0## (2). The derivation itself I understood quiet good, but what I don't understand is where the equation (1) came from, because in my research it was just given and not derived from anywhere. Does anybody know where (1) comes from or why from it the...

Similar threads

Replies
17
Views
2K
Replies
10
Views
2K
Replies
2
Views
485
Replies
1
Views
2K
Replies
3
Views
5K
Replies
11
Views
4K
Replies
3
Views
3K
Back
Top