I am trying to measure a radio TX power that is plugged directly to my spectrum analyzer with a -40db attenuator to compare to what the manufacture claims it be and I have few question.

The manufacture claims that the RF power is 30dBm regardless of the bandwidth that is selected. The bandwidth can be selected from 1MHz to 8MHz at frequency from 1.8GHz to 1.85GHz. I understand with higher bandwidth my sensitivity decreases.

However, I am not measuring 30dBm on all BW or I am measuring it wrong so I need some clarification of how exactly to interpret the data I have gathered.

I have tested that radio in all 4 BW (1MHz, 2MHz, 4MHz, 8MHz) all have the TX power output selected at 30dbm. With 40db attenuator I know that I should see -10dbm with some losses in the cable.

Here is the results I got from setting the spectrum analyzer to hold max value and looking for peak:

@1MHz BW: -10dbm @2MHz BW: -12dbm @4MHz BW: -14.9dbm @8MHz BW: -17.8dbm

My question is, when they claim a 1W radio, do they look for a peak of the signal or is it an integration of the entire bandwidth that is 1W. Because if it is the latter then the radio might be as advertised. Since the area under the curve most likely is the same since with higher bandwidth the average power looks the same (I didn't do any math to verify this I am just assuming based on what I saw on the spectrum output)

However, if it is the peak power that is supposed to be 30dbm regardless of the bandwidth used and we just lose receive sensitivity with higher bandwidth then perhaps something is not right here.

The manufacture is a small company and I have been having difficult time getting in contact with their engineers so that is why I am here hopping some one that has had some experience in this kind of thing that can shed some light on it.



1 Answer 1


My question is, when they claim a 1W radio, do they look for a peak of the signal or is it an integration of the entire bandwidth that is 1W.

Power is necessarily the integral of all of the output across frequency.

A helpful way to think about it, is if you transmitted into a resistor, how much would you heat it up? Some broadband power meters actually work that way.

  • \$\begingroup\$ Okay so you are saying that this is how it is intended and everything works as advertised? \$\endgroup\$
    – J. Jones
    Oct 9, 2020 at 16:55
  • \$\begingroup\$ That's definitely how it is defined, and your measurements seem moderately consistent with that \$\endgroup\$ Oct 9, 2020 at 16:56
  • \$\begingroup\$ One more question, when I do my path loss calculations, do I use peak power out or total power? \$\endgroup\$
    – J. Jones
    Oct 9, 2020 at 18:10
  • \$\begingroup\$ Presumably you would use whatever you care about; eg, you'd have to consider how the criteria of what you need after the loss is defined. Also you have to consider the receiver bandwidth: a wider filter will pick up more noise, too, because the noise is in dB/Hz and you have to integrate over the filter bandwidth to get the noise power making it through the filter. \$\endgroup\$ Oct 9, 2020 at 18:13

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.