# How do I compute the average power from the PSD plot attached?

I want to calculate the average Power from the Power spectral density plot I have here. Any help would be greatly appreciated.

• Is there no option that calculates this on your average value? I've never used this measuring device but surely should be an averaging calculation feature. I don't think it's possible to calculate a generated PSD without knowing $F(\omega)$, per se.
– user103380
Commented Jun 21, 2019 at 16:42

There's no simple, accurate way to do it from a graphical plot.

The two alternatives are (a) simplest, set the RBW to 3 kHz, which will encompass essentially all the power in the signal, now read the marker

or (b) hard work, get a numerical dump of the display, and use anything from Excel to your favourite computer language to integrate the power across the display.

However, by eyeball, the marker is at -51dBm in a 300Hz RBW. The power has dropped by 10dB by +/- 500Hz from the centre, and you get precious little extra power after that. So your total power will be slightly more than -51dBm, maybe -50dBm, maybe -49dBm.

You simply increase RBW to include the spectrum for which to want the PSD.

Estimate what it your think it should be by this graph , before you increase it towards the VBW=100kHz to exclude noise to get about -50 dBm @3kHz or 10kHz and see how close yours and my estimate is.

You can also go to zero span at centre f with RBW=30kHz and compute from the time trace the Average and the peak power and compare with the PSD Power for your education on quasi-peak detection of PSD results.

Any good SA or external output to a DSO can compute any of these from the zero span matched signal BW time trace output.