All Answers (27)
 Power spectral density can be obtained computing autocorrelation function in Fourier space. From physical point of view, high frequency wavenumber of the Power Spectral Density of a 2D image, can be related to sharp variations in intensity values of the pixels in the image domain. Such variations occur in pixels near to an object edge in the image.
 Power Spectral Density is the Fourier Transform of the autocorrelation function of a signal. First compute the auto correlation function and then compute its Fourier Transform.
 White noise will give a flat power spectral density (White)
 Autocorrelation in Fourier space means to me Fourier Transform of Autocorrelation function.
 It does not look obvious when you say "Auto correlation in Fourier Space" to a person who does not know what it is about.
 Ok, thanks for clarifying my answer!
 If you get into the computation of the Fourier Transform of the auto correlation funciton, you will find that you can do a 2sided or a 1sided Fourier Transform and they both give different results. The 2sided Fourier Transform of the ACF is called a correlogram and the 1sided Fourier Transform of the ACF is periodogram. Usually, the correlogram is more accurate than the periodogram. A good book to learn the applications of this would be Spectral Analysis of Signals by Petre Stoica and Randolph Moses. Usually the EEG of nerve signals gives strong crosscorrelations during neural activity and weak crosscorrelations during lack of neural activity. This property is used in prosthetic devices to identify the activity the brain wishes to perform.
 For a 2sided Fourier Transform (correlogram), the index at 0 delay is taken as the midpoint and the summation goes from N/2 to N/2.
 I hope that clarifies things better.
 Here is a good EEG data set I could find online: http://sccn.ucsd.edu/~arno/fam2data/publicly_available_EEG_data.html I think I might be spending some time trying to get some results on this.
 Ok....thanks to both of you...ultimately our target is to convert any signal to its sine and cosine term, as in nature all signal are sinusoidal, psd is an anathor form or i can say a kind of representation....m i correct????
 Hi Nisha. The Fourier Transform theorem states that only periodic signals have a Fourier Series representation. When we do a Discrete Fourier Transform, we are assuming the signal to be periodic outside the sampling period. It does not make any errors in the process and is the function mapping can be inverted to get the original signal back.
 Thanks to all...
 What are negative frequencies? Do they really exist or used in mathematical formulas only?
 Negative frequencies are required to complete Fourier integral. Negative frequencies correspond to the negative time. They are also called mirror frequencies : you cannot get (modify) your mirror image as you cannot modify these frequencies as you can get negative time (= to modify pass time or action).

Power spectral density describes the signal power distribution over the frequency.

Power spectral density is distribution of power, and it can be calculated by Fourier Transform of autocorrelation function of the signal. You can test this to better understand. Take a signal/image, find autocorrelation and take FFT, plot it.

Power spectral density function (PSD) shows the strength of the variations(energy) as a function of frequency. In other words, it shows at which frequencies variations are strong and at which frequencies variations are weak. The unit of PSD is energy per frequency(width) and you can obtain energy within a specific frequency range by integrating PSD within that frequency range. Computation of PSD is done directly by the method called FFT or computing autocorrelation function and then transforming it.

The power spectral density shows how the energy of a signal is distributed. That can allow us to make some decisions or operations on a considered signal. The important thing is that the DSP can be obtained from the function of auto correlation without signal priory knowledge (WienerKhintchine theorem 1929). What makes the valid FT for a signal is random or real.

Hi
I advice you to see these documents. You will find what you need with some examples.
Best regards

when you measure the signal it will be having the various frequency components in it, since the power contained in each signal with different frequency will be different so the psd is the graph of power vs frequency.
also the main application of the psd plot is that it will give the range of the frequency where the power density is high
hope this will satisfy your question

Power spectral density function (PSD) shows the strength of the variations(energy) as a function of frequency. In other words, it shows at which frequencies variations are strong and at which frequencies variations are weak. The unit of PSD is energy per frequency(width) and you can obtain energy within a specific frequency range by integrating PSD within that frequency range. Computation of PSD is done directly by the method called FFT or computing autocorrelation function and then transforming it.
Go here for more...

Power spectral density function shows the energy with frequency. It will useful when you want check the signal is week or strength.

What is the difference between a PSD = f( Hz) and the square of energy (E^2) = f(Hz)?
When a domain in Hz is changing in energy (increase of E^2) maintained along all the signal following. In other words what is the difference in the meaning of density (PSD) versus only the square of E (E^2) = Hz?
What is lost in the information relatively 
Hi
Power Spectral Density ou PSD is the square of the Fourier transform module, divided by the integration time T (or, more strictly, the limit as t goes to infinity of the mathematical expectation of the square of the Fourier transform module signal  this is called average power spectral density). Thus, if x is a signal X and its Fourier transform, the spectral power density is:
PSD=ІXІ^2/T
Best regards

PSD means how the strength of a signal is distributed in the frequency domain.
 Views 41325
 Followers 37
 Answers 27
Popular Answers
Power spectral density describes the signal power distribution over the frequency.