-3

So here and here,

RF power envelope

We have determined the information expressed in an AM signal comes from the fact that the amplitude of electromagnetic energy at any given frequency determines the amplitude of sound coming from the speaker. The fact that there is no energy at a particular frequency is part of the equation. The lack of energy in one part and the emphasis of energy in other parts of the bandwidth encodes the information

This makes AM very subject to broadband impulse energy, lightning strikes, etc. because the receiver cannot discriminate between that noise and the signal transmitted by the sending station.

So then We have FM. Here and here What is the basic difference between AM and FM radio?

So, my question.

If we used to encode information by varying the amplitude of a range of RF frequencies, and now we encode information by just varying whether a constant amplitude frequency is present in the RF energy

Aren't we just saying that we determined the amplitude changes to be a noisy encoding channel, and we quit using them? Isn't this wasting information carrying capacity? We used to encode information there.

**UPDATE Based on the comments below, I had another epiphany. FM is basically an error correcting protocol. Just like a CRC , that wastes information capacity due to redundancy, but ensures message correctness. FM wastes capacity, but ensures correctness.

http://en.wikipedia.org/wiki/Cyclic_redundancy_check

Community
  • 1
Andyz Smith
  • 395
  • 2
  • 12
  • Note also that in AM, you can not only suppress one of the side bands, but also the carrier. The bandwidth needed by AM is basically that of the baseband signal being encoded. FM intermodulation products go out to infinity. – Kaz Aug 24 '13 at 14:53
  • Not practically though. A instantaneous shift from one frequency to another in an FM envelope will cause infinite power in infinitely many higher order harmonics, but once you get past the first harmonic the power is so small as to be practically nonexistent. – Andyz Smith Aug 24 '13 at 14:55
  • I'd rephrase your opening paragraph along the lines that the originating baseband signal (due to AM) is shifted up in the spectrum and after reception by a radio receiver (and appropriate filtering to remove unwanted channels close by) the receiver's demodulator shifts it back down so it becomes audio again. – Andy aka Aug 24 '13 at 14:56
  • Yes, so this seems like the original naive way of encoding, selected only because the electronics involved were simple. Once we figured out that transposing information into RF amplitude we discovered that, Just like you need a quiet room to listen to classical, you need a quiet RF environment to transmit information this way. – Andyz Smith Aug 24 '13 at 15:00
  • kaz, Wikipedia , Carson Rule, ... ... **Any** modulated signal will have an infinite number of sidebands and hence an infinite bandwidth but in practice all significant sideband energy (98% or more) is concentrated within the bandwidth defined by Carson's rule. It is a useful approximation, but setting the arbitrary definition of occupied bandwidth at 98% of the power still means that the power outside the band is only about 17 dB less than the carrier – Andyz Smith Aug 24 '13 at 15:15
  • 1
    NO, there can be only two sidebands. NO, not an infinite bandwidth for AM. YES, a theoretically infinite bandwidth for FM. NO, FM is NOT "basically an error correcting protocol". YES, FM wastes capacity (compared to AM) and creates the opportunity to avoid common noise sources and even close-by transmissions. FYI "Frequency Deviation" may seem an attractive choice of words to apply to AM but it's best left describing FM. – Andy aka Aug 24 '13 at 16:34
  • I maintain, AM and FM both have theoretically infinite number of infinite small by products of modulation. – Andyz Smith Aug 24 '13 at 17:40
  • As well, frequency deviation must be used to describe AM bandwidth. – Andyz Smith Aug 24 '13 at 17:41
  • FM frequency deviation is equivalent to bandwidth. ( yes ) AM bandwidth is equivalent to frequency deviation? ( yes AM bandwidth is from x kHz to y kHz ) All the same. – Andyz Smith Aug 24 '13 at 17:55
  • No, AM has only two sidebands. Check the mathematics. FM maximum frequency deviation is equivalent to FM bandwidth but it encodes signal amplitude. AM frequency deviation is +/- the allowed audio bandwidth, and it encodes signal frequency. Not the same thing. – user207421 Jul 08 '14 at 03:38

1 Answers1

1

The basic difference between AM and FM is that for AM the modulating signal changes the amplitude of its carrier frequency while it changes the frequency in FM. Generally a modulated FM signal will occupy more bandwidth than an AM signal. In that respect, it is less efficient with respect to bandwidth. The tradeoff is that the higher bandwidth results in a demodulated signal with a higher SNR than AM. Note that if you modulate the amplitude of a carrier with a sine wave, the resultant AM signal has only 3 components, the carrier and a sideband 1 kHz above the carrier and another 1 kHz below. It is not infinite. On the other hand, doing FM modulation with the same sine wave theoretically results in an infinite bandwidth since there are sidebands spaced 1 kHz apart on both sides of the carrier. The amplitude of these sidebands decreases as their separation from the carrier increases so, in practice, the bandwidth of an FM signal can be band limited and the modulating signal can still be recovered with little distortion (hence the success of commercial FM broadcasting). Because FM relies on frequency and not amplitude, it is less susceptible to many forms of interference encountered in radio transmissions such as lightning strikes. Most FM receivers, in fact, amplify and then clip the incoming received signal in order to lessen the effect of amplitude noise since all of the modulation information is contained in the zero crossings of the signal and not its ampllitude. The main point I want to convey is that the main advantage of FM over AM is its ability to trade off increased bandwidth for increased signal to noise ratio. This tradeoff occurs in many information transmission systems including digital.

Barry
  • 15,733
  • 1
  • 26
  • 28
  • I'm still digging on this. On AM , When you say 'modulate the carrier' you mean, modulate the frequencies around the carrier. – Andyz Smith Aug 24 '13 at 23:27
  • Modulating the carrier for AM means creating the following signal: (1+mg(t))sin(2πft) where m is the modulation index (goes from 0 to 1), g(t) is the modulating signal, and f is the carrier frequency (t is time). The resulting signal is basically a sine wave whose amplitude varies in step with the modulating signal (g(t)). – Barry Aug 25 '13 at 00:25