Phase Spectra from Fourier Transform

In summary, to read the phase spectra from a Fourier Transform, you need to take the arctangent of the ratio of imaginary to real components, rather than just looking at the position of the vector on the complex plane. This will give you the correct phase for each frequency component.
  • #1
frenzal_dude
77
0
How can you read the phase spectra from a Fourier Transform?

if [tex]g(t) = Sin(2\pi f_{c}t)[/tex]

then for the single sided spectrum, you have one frequency component at [tex]f=f_{c}[/tex] with a height of [tex]\frac{1}{j}[/tex] which from looking at the complex plane, corresponds to a phase of [tex]\frac{\pi }{2}[/tex] (ie. [tex]g(t) = Sin(2\pi f_{c}t)[/tex] is made up of a cosine component with [tex]f=f_{c}[/tex] and phase = [tex]\frac{\pi }{2}[/tex].

But, if you consider [tex]\frac{1}{j} = -j[/tex], then the phase would correspond to [tex]\frac{3\pi }{2}[/tex] which would in effect negate the amplitude ([tex]Cos(x - \frac{3\pi }{2}) = -Cos(x - \frac{\pi }{2})[/tex].

So which complex amplitude should be considered correct?[tex]\frac{1}{j} or -j[/tex] ?
 
Engineering news on Phys.org
  • #2
frenzal_dude said:
How can you read the phase spectra from a Fourier Transform?

if [tex]g(t) = Sin(2\pi f_{c}t)[/tex]

then for the single sided spectrum, you have one frequency component at [tex]f=f_{c}[/tex] with a height of [tex]\frac{1}{j}[/tex] which from looking at the complex plane, corresponds to a phase of [tex]\frac{\pi }{2}[/tex] (ie. [tex]g(t) = Sin(2\pi f_{c}t)[/tex] is made up of a cosine component with [tex]f=f_{c}[/tex] and phase = [tex]\frac{\pi }{2}[/tex].

But, if you consider [tex]\frac{1}{j} = -j[/tex], then the phase would correspond to [tex]\frac{3\pi }{2}[/tex] which would in effect negate the amplitude ([tex]Cos(x - \frac{3\pi }{2}) = -Cos(x - \frac{\pi }{2})[/tex].

So which complex amplitude should be considered correct?[tex]\frac{1}{j} or -j[/tex] ?

I've worked it out (for anyone whose found this in a google search or something).

To find the phase spectra you need to take the arctangent of the ratio of imaginary to real components, and NOT merely look at the position of the vector on the complex plane.

So first you need to get those imaginary components to have j on the numerator instead of on the denominator, that way you know the coefficient of it.

So for Sin(sPIfct) the impulse at f=fc has an imaginary coefficient of -1/2, so the ratio of imaginary to real is -infinity (since there is no real part, real=0), arctan of -infinity = -PI/2 radians.

For the impulse at f=-fc, the coefficient of the imaginary component is 1/2, the ratio is +infinity, arctan of +infinity = +pi/2 :)
 

1. What is a phase spectrum?

A phase spectrum is a graphical representation of the phase shift or delay of a sinusoidal signal at different frequencies. It shows the relationship between the input signal and the output signal in the frequency domain.

2. What is a Fourier transform?

A Fourier transform is a mathematical tool that breaks down a signal into its individual frequency components. It converts a time-domain signal into its equivalent representation in the frequency domain.

3. How is a phase spectrum obtained from a Fourier transform?

To obtain a phase spectrum from a Fourier transform, the phase shift of each frequency component is calculated and plotted against the corresponding frequency. This can be done using mathematical formulas or by using software that performs the Fourier transform.

4. Why is phase information important in signal processing?

Phase information is important in signal processing because it provides insight into the timing and synchronization of different frequency components in a signal. This can be useful in applications such as audio and image processing, where preserving the phase relationship between different frequency components is crucial for maintaining the integrity of the signal.

5. Can a phase spectrum be used to reconstruct the original signal?

No, a phase spectrum alone cannot be used to reconstruct the original signal. It only provides information about the phase shift at different frequencies, but not the amplitude or time-domain values of the signal. The original signal can only be reconstructed if the amplitude and phase information are both known, which is why the Fourier transform is typically used in conjunction with the inverse Fourier transform to convert a signal back to the time domain.

Similar threads

  • Electrical Engineering
Replies
2
Views
1K
  • Calculus and Beyond Homework Help
Replies
5
Views
328
Replies
3
Views
530
Replies
5
Views
1K
  • Electrical Engineering
Replies
4
Views
799
  • Calculus and Beyond Homework Help
Replies
1
Views
195
Replies
4
Views
276
  • Calculus and Beyond Homework Help
Replies
6
Views
2K
  • Differential Equations
Replies
4
Views
2K
  • Calculus and Beyond Homework Help
Replies
1
Views
330
Back
Top