How to Make Channel Power Measurements

Learn how to calculate the power level of a spread spectrum signal.

Check out the EEs Talk Tech electrical engineering podcast:

The Keysight Podcasts YouTube channel:

Transcript
You might be used to measuring power with a marker placed on a frequency peak, but that isn’t going to work for spread spectrum signals.

Hi, I’m Ally, and welcome to Ready, Set, Measure. Today, we’ll learn how to measure channel power for spread spectrum signals.

To measure the power of a continuous wave signal, all you have to do is add a marker at the peak of the signal. But, this doesn’t work for signals like this WCDMA signal.

Since the WCDMA signal is spread across a spectrum, the marker mistakenly reads off the power of the signal to be about minus 20 dBm, but we know the signal power is actually closer to 0 dBm. In this case, the marker tool isn’t reliable.

To find the power of this modulated signal we need to find the area under the curve, in other words, we need to integrate. But, an integration over this many frequencies, manually, is tedious and can be error prone.

If you don’t want to do it manually, you can use a signal analyzer’s built-in channel power measurement.

To be able to do this, all we have to do is input the center frequency and the integration bandwidth of our signal. Then the signal analyzer’s channel power measurement automatically calculates the total channel power and power spectral density.

So, in summary, if you want to measure the power of a spread spectrum signal, you either have to do manual calculations or use a signal analyzer with a channel power measurement.

#engineering #howto #rfmeasurements #rftest #signalanalyzer #signalanalysis #rf

[manideepdunna]

I am guessing it works only for continuously available WCDMA signals. Is this channel power measurement applicable for bursty WCDMA signals as well? Does it give the average power measured for the whole burst period?

[RobeenaShepherd]

Integration can be tedious and error prone, especially if like me you haven't done it for an embarrassingly long time and can't even remember how to do it, even though you have a piece of paper that tells you once upon a time you definitely could do it!

[dcdeFPV]

I can understand the error-prone part, but are manual calculations less accurate simply because they're not done at as high of a resolution? i.e. the human just sees what's on the screen and might have to estimate values a bit, where as the machine can "zoom in" much farther into the raw data? Also, a livestream/product demo where youtubers could ask product questions in chat and see features demonstrated in real time would be really cool. Not sure if that's ever been done or considered. Anyway, thanks for the video, informational as always!

[misbahpatel]

Hello, I have a query regarding ow to increase the signal power of function generator upto 60-80Mhz usin AMPLIFIER?

[natezimmer]

Why does a peak hold reading not accurately reflect power for a spread spectrum signal? I can see why the average might be lower for a spread spectrum signal but i'd think the peak / instantaneous power would be the same. What am I missing?