spectral analysis, frequency spectrum, power spectral density

I have power=2000 float data points and time= 2000 float time intervals. in 200ns.

is there any way i can put this in matlab to see some spectral analysis of frequency and other things. basically i have x and y values of power vs time. I want to manuplite this data to show frequency values which i expect to be in 500-1000MHz range.

or may be draw a fft plot. but the problem is i just know these two variables (power and time) is it possible? if yes then how?

[Walter Roberson]

Are the times evenly spaced or irregularly spaced?

[hardik]

time is evenly spaced since we are using a sampling rate of 10e9.so yea, but the amplitude keeps on changing

Answer 1

[Rick Rosson]

   N = 2000;
   StartTime = 0;
   StopTime = 200e-9;
   dt = (StopTime - StartTime)/N;
   Fs = 1/dt;
   dF = Fs/N;
   f = -Fs/2:dF:Fs/2-dF;

Answer 2

[Rick Rosson]

   >> doc freqz

[hardik]

thanks for ur answer....it helps a lot but some clarity i need...when i do freqz(w) for the power data. i see x values of freq as normalised freq as api rad/sample. can i change this to actual frequency. say like in MHZ?

I have a sample of power values generated in 200ns. this values of power vary with different frequency. with a high freq the power goes very high. and then drops back to low freq. its about partial discharge.

so i was wondering if i can see the frequency values as well with this.?

Answer 3

[Wayne King]

"thanks for ur answer....it helps a lot but some clarity i need...when i do freqz(w) for the power data. i see x values of freq as normalised freq as api rad/sample. can i change this to actual frequency. say like in MHZ?"

Have you read the documentation for freqz()? It shows that you can use the sampling frequency as an input argument

 [H,F] = freqz(B,A,N,Fs);

For example:

      Fs = 1e4;
      % Butterworth filter with 3dB frequency of 1 kHz
      [B,A] = butter(10,(2*1e3)/Fs); 
      [H,F] = freqz(B,A,[],Fs);
      plot(F,20*log10(abs(H)));
      grid on; xlabel('Hz');