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 auto-correlation function of the signal. You can test this to better understand. Take a signal/image, find auto-correlation 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 (Wiener-Khintchine theorem 1929). What makes the valid FT for a signal is random or real.
I advice you to see these documents. You will find what you need with some examples.
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
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
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 means how the strength of a signal is distributed in the frequency domain.
University of Bergen
National Polytechnic School of Algiers
Indian Institute of Technology Bombay
Indian Institute of Technology Delhi
Ng Boon Sing
Advanced Materials Technologies Pte Ltd, Singapore
S. Bali Reddy
Indian Institute of Technology Guwahati
Taha A. Ali
Arab Academy for Science, Technology & Maritime Transport
Universidade Federal do Rio Grande do Sul
Tarbiat Modares University