0
\$\begingroup\$

I read (for instance here) that FM modulation is less affected by noise than AM modulation.

But I don't understand conceptually the reason for that: why is the SNR with frequency modulation higher than the SNR with amplitude modulation?

\$\endgroup\$
  • \$\begingroup\$ Re-read page 3 of the PDF you linked to. \$\endgroup\$ – JRE Jul 1 at 20:40
  • \$\begingroup\$ @JRE: That's just a bunch of hand-waving. Doesn't really explain anything. \$\endgroup\$ – Dave Tweed Jul 1 at 20:42
  • \$\begingroup\$ FM responds to FM and rejects AM modulation or superposition of noise. To some extent due to CNR to SNR demodulator improvement factor using more BW on carrier than Audio ( depending on band,subband). Also Lightning and engine noise does not reach 100MHz band ( highly attenuated). The real difference is due to the detectors. AM uses linear amplitude detectors ad FM uses linear freq mixers (multipliers) and often PLL demodulation of carrier shift \$\endgroup\$ – Sunnyskyguy EE75 Jul 1 at 20:44
  • \$\begingroup\$ Carson’s rule for wideband FM means there is a log effect of CNR/SNR improvement not possible in AM. Due to delta f/f >>1 \$\endgroup\$ – Sunnyskyguy EE75 Jul 1 at 20:51
  • \$\begingroup\$ I'm voting to close this question as off-topic because this is electrical engineering, but very specific – chances are high that signals.stackexchange.com is a more appropriate place to ask this. (recommending migration there) \$\endgroup\$ – Marcus Müller Jul 1 at 21:34
0
\$\begingroup\$

In a radio receiver, noise will generally appear as variations in the amplitude of the signal, and will be processed by the AM demodulator in the same way as the desired signal, so will be heard by the user.

In an FM receiver, the received signal is limited or clipped to remove amplitude variations (such as noise), and the FM demodulator will generally ignore any amplitude variations that get to it, so the noise on the initial signal will not be heard by the user.

\$\endgroup\$
  • 1
    \$\begingroup\$ And what about phase noise? Why is it less than amplitude noise? In theory it will affect very much the FM signals \$\endgroup\$ – Kinka-Byo Jul 2 at 1:50
  • 1
    \$\begingroup\$ random-phase noise is well-modeled with 50% of the noise power being AM, and the other 50% of power being PM which modulates the zero-crossing timings. \$\endgroup\$ – analogsystemsrf Jul 2 at 2:11
  • \$\begingroup\$ So is the advantage of FM the fact that it sees only the second one, while AM sees both? \$\endgroup\$ – Kinka-Byo Jul 2 at 5:41
0
\$\begingroup\$

IN A VERY SIMPLIFIED WORLD. It is easier to "segragate" noise on the frequency domain because [white] noice is relative small across the frequency spectrum compared to the carrier sinal. Hense, you only pay attention to the frequency of the signal with the highest power, and track it. So at the end, noise adds-up in a non-desctructive way on your FM signal because it does NOT interfere with your tracking [ or at least, as long as your carrier's power is much higher than the noise's power].

\$\endgroup\$
0
\$\begingroup\$

Signal to noise goes as you'd expect, but in FM it's rather more subtle.

Let's say a transmission has white noise added to it in transit. That noise is uncorrelated to the signal, so half of it looks like AM (in phase with the carrier) and half of it looks like PM (phase modulation).

If the want to improve the SNR due to AM, then we can increase the carrier power of our transmitter. This gives us a linear improvement in SNR at the receiver, at the cost of real transmitted power. This is measured in watts, and so money for bigger transmitters, more power for them etc. If the interference is from natural sources, then this works. If the interference is from other transmitters further way, then 'everybody shouting louder' does not work.

If we want to improve SNR due to PM, then we can increase the carrier deviation of our transmitter. It costs no more in watts to send a signal that has +/- 10 radians deviation than one that has +/- 1 radian deviation, however the first one will have 20dB better SNR. You don't get owt for nowt, and the cost of increasing the deviation power of the transmitter is that you can cram fewer wide deviation transmitters into any given bandwidth than AM transmitters.

\$\endgroup\$

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.