- #1
Sklar Hast
- 1
- 0
Hi...
I have several Power Spectrua that I want to average into a single representative power spectrum. Then, I have to analize the amplitude spectrum of the average (i.e. sqrt(power spectrum))
Doing that doesn't give the same result as averaging the amplitude spectrums.
Supose I have tho spectrums that I want to average, each having one datapoint , whose power are A and B, respectively. a and b are the square roots of A and B, so
a^2=A
b^2=B
Method 1:
sqrt{ (A+B)/2 }= sqrt{ (a^2+b^2)/2 }
Method 2:
(a+b)/2
I suspect that I'm coliding with Schwarz innequality (or triangular inequality)
What do you recommend me to progress in my ferequency domain analisis?
Thanks
I have several Power Spectrua that I want to average into a single representative power spectrum. Then, I have to analize the amplitude spectrum of the average (i.e. sqrt(power spectrum))
Doing that doesn't give the same result as averaging the amplitude spectrums.
Supose I have tho spectrums that I want to average, each having one datapoint , whose power are A and B, respectively. a and b are the square roots of A and B, so
a^2=A
b^2=B
Method 1:
sqrt{ (A+B)/2 }= sqrt{ (a^2+b^2)/2 }
Method 2:
(a+b)/2
I suspect that I'm coliding with Schwarz innequality (or triangular inequality)
What do you recommend me to progress in my ferequency domain analisis?
Thanks