Green Xenon [Radium] wrote:
(snip)

> Yeah but even if the signal being transmitted/received
> is digital, it's carrier wave is still analog. Right?

> AFAIK, there is no such thing as a digital carrier wave. The carrier
> wave is always analog just like a cable link is also always analog. The
> signal transmitted through the analog medium maybe digital, though.

> If a PCM signal [digital] is transmitted on an AM carrier wave [analog],
> the AM wave's peak-to-peak amplitude will vary according to the PCM
> signal in the following manner:

> 1. A positive amperage of the PCM signal will cause the AM carrier
> wave's peak-to-peak amplitude to increase while a negative amperage
> [i.e. going below the x-axis when graphed] will cause a decrease the AM
> carrier's peak-to-peak amplitude.

That is one way. Usually you will also need some synchronizing
information to know which bits are which. Sometimes it is done
as a subcarrier, modulating a carrier synchronous to the bit rate,
and then using that to amplitude modulate the real carrier.

> 2. A increase in frequency of the PCM signal will cause the AM carrier's
> peak-to-peak amplitude to vary more rapidly while a decrease in the PCM
> signal's frequency will cause the AM wave's peak-to-peak amplitude to
> vary more slowly.

> That's how I understand it. The PCM signal is digital but causes a
> measurable affect on the analog AM carrier wave. Upon reception, an
> AM-demodulator can retrieve this PCM signal and play it back. I could be
> wrong though.

You could also generate a pulse for 1 and no pulse for 0, though
there are synchronizing problems with that. Next easiest is
a pulse for 1, no pulse for zero, and a clock pulse half way
in between. (The coding used on single density floppy disks.)

You need some kind of modulation for an AC coupled signal path.

-- glen