Is it possible to digitally extract received power info (like VU meter perhaps) from and PSK31 audio (or alike), actually the mode for text encoding over audio hasn't been chosen yet and the capability (or not) to extract power info from it would be a key factor, so suggestions on this topic would also be appreciated ;)

PS: this is intended in the VHF 140-160 MHZ band


Assuming you are doing something like recording the audio output from an SSB radio, then the audio recording is just the RF spectrum shifted down in frequency. So, you could calculate the RMS average of the audio file, square it (since power is proportional to the square of voltage), and you'd have some measurement of power.

However, this requires that the gain between the antenna and the audio recording is some fixed, known value. If your radio has AGC, this is certainly not the case, since the AGC will adjust the gain to produce audio recordings of approximately the same amplitude, regardless of the received power.

If you just want some measure of signal quality, there are other methods you could use which do not have this requirement. For example, you could estimate the signal-to-noise ratio. Alternately, if you use an encoding with forward error correction (PSK31 has none, but some variants do) you may have a means to estimate the bit error rate.

| improve this answer | |
  • $\begingroup$ Although most of the VHF Radios we'got are FM ones, a few of them are SSB, so it might worth testing with them using your suggested method $\endgroup$ – Gonzalo Vasquez Jan 2 '17 at 21:28
  • $\begingroup$ I think your third paragraph will work for me but, as I'm not an expert on such encodings, could you please provide a reference link on what encodings do offer forward error correction? $\endgroup$ – Gonzalo Vasquez Jan 23 '17 at 18:54
  • $\begingroup$ @GonzaloVasquez FEC is very common so it would be a long list. Your best bet is to research a particular modulation and encoding and see what it offers. You might look at D-star data mode for a start: unlike D-star voice mode it's free and open, while retaining the benefit of being well adopted. I think it includes FEC. $\endgroup$ – Phil Frost - W8II Jan 23 '17 at 19:40

That doesn't depend on the text-to-audio modulator, but on the audio-to-baseband / audio-to-RF modulator.

With AM, audio amplitude is proportional to received signal strength, with FM, the received signal strength has absolutely no effect on audio signal (and hence, you can't recover the received power at all; that info is simply lost on the way).

In any case, I think you're not measuring the right thing – if you're really concerned about signal quality, you'd rather go for a measure of SNR in the audio prior to text de-coding. Received power is a meaningless figure without knowing how your receiver reacts to that, how much of that power is noise or interference, and how sensitive your audio demodulator is to that, and what the effects of audio distortions on your text communication are.

So, I personally think you might want to ask yet another question on ham.stackexchange.com: Given that I want to achieve X, how can I choose a digital/text data codec? and describe what that X and its environment is.

| improve this answer | |
  • $\begingroup$ Most our hardware is FM, but SNR does make sense to me $\endgroup$ – Gonzalo Vasquez Jan 2 '17 at 21:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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