I would like to detect the phase (only) of a standard FM transmitter carrier over the air while it is modulated, while at the same time avoiding receiver threshold effect.

Typical commercial FM transmitters use +/- 75 kHz carrier deviation and modulating frequencies of 15 kHz or higher. I need to lock to the FM carrier and track it while it is modulated, and recover just the steady-state carrier phase. I'd also like to avoid the usual 10 dB threshold effect in order to achieve maximum sensitivity under weak signal conditions.

Can I do all this using a PLL detector and an SDR? If I understand correctly, the loop bandwidth would be made narrow (say, 1 Hz), therefore removing the modulation and leaving only the carrier itself. Eliminating the limiter should eliminate the threshold effect and allow the PLL to track the carrier down into the noise for maximum sensitivity.

Any thoughts or suggestions will be appreciated.

  • \$\begingroup\$ Modulating frequencies of 20Hz-53kHz actually. \$\endgroup\$ – Marquis of Lorne May 16 '15 at 21:38
  • \$\begingroup\$ And a 57kHz subcarrier for RDS data (where applicable) \$\endgroup\$ – Brian Drummond May 16 '15 at 22:59
  • \$\begingroup\$ And another at 67kHz for SCA transmissions, if anybody is still doing that. \$\endgroup\$ – Marquis of Lorne May 17 '15 at 10:22

Your proposal is to measure the long-term average phase of a broadcast FM signal by stripping it of modulation in a PLL. This is realizable and similar to what the AFC feedback in an FM receiver already does. The off-air frequency information can be useful as a reference (since regulators require broadcasters to keep within specific frequencies) but phase information only adds the propagation delay from transmitter Tx to your receiver Rx. You may be thinking of using very directional antenna(s) to compare the delays along multiple paths from Tx to Rx as some kind of secondary radar or navigator, in which case any distances calculated will be modulus the wavelngth which is about 3 meters at 100 MHz.

The PLL detector must be a linear mixer where the input phases lock in quadrature, not the edge-counting type that aims for 0 degree phase lock. CMOS IC type 4046 contains both kinds of phase detector but needs to work at an IF no more than a few MHz. This is because a characteristic of FM is that the carrier vanishes at some sinusoidal modulation depths, and that may also happen occasionally on broadcast material.

| improve this answer | |
  • \$\begingroup\$ Excellent points. But in your second paragraph didn't you mean that the carrier <b>component</b> in the Bessel expansion vanishes at certain mod indexes but the carrier seen in the time domain remains constant amplitude? Since my objective is to continuously track the carrier to derive its average phase, that would be impossible to do if the carrier itself were to disappear under certain modulating conditions. Also, I don't understand how the choice of IF applies here? \$\endgroup\$ – Palapaguy Jun 4 '15 at 22:39

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.