SCREEN-L Archives

April 1991

SCREEN-L@LISTSERV.UA.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Jeremy Butler <[log in to unmask]>
Reply To:
Film and TV Studies Discussion List <[log in to unmask]>
Date:
Sun, 21 Apr 91 21:20:35 CDT
Content-Type:
text/plain
Parts/Attachments:
text/plain (110 lines)
This article is fairly long and somewhat technical, so you should feel
free to skip it if you care to take my word that FM does *not*
transmit a change in frequency based on a difference in source from
moment to moment.  I'm posting this because I don't like to see
completely incorrect information flying about.
 
In article <[log in to unmask]>
  82 Malcolm Dean 213-5-5676 <[log in to unmask]> writes:
 
> > ...FM does transmit a signal directly, rather than
> > any difference.  Both FM and AM radio modulate a carrier to transmit
> > their signal, an AM transmitter changes the amplitude of the carrier
> > in proportion to the input signal, an FM transmitter changes the
> > frequency of the carrier in proportion to the input signal.
>
> This is not a forum in radio theory, but your explanation
> confirms what I said, which is that FM transmits a change in
> frequency reflecting the difference in the source from moment to
> moment.
 
Nope.
 
The following explanation is pretty simple.  I'll just go over some
basics first to make sure that nobody gets lost.
 
When we convert sound to an electrical voltage using a microphone
we come out with a signal that varies in amplitude (level) with
time.  This amplitude difference is measured in dB.  (A dB is a
measure just like a centimeter or a gram, but it measures the
voltage or power of an electrical or sound signal based on an
agreed-upon reference level.)  This changing electrical signal can
be sent to an FM transmitter directly or stored on a tape or a
record or a CD or whatever and sent to a transmitter at a later
time.
 
Now, the FM transmitter always transmits a signal at the same level.
The way it describes these level differences in the input, which is
the information that we want to transmit, is though a corresponding
change in frequency of the signal.  The receiver looks at the
frequency of the signal coming in, changes it back to an amplitude,
and sends it out to an amplifier, which puts it out a speaker.
 
(Incidentally, this need to convert a frequency to an amplitude is why
an FM receiver is more complex than an AM receiver.)
 
Now, our current FM system transmits a frequency that is
*directly proportional* to the *absolute* value of the source from
moment to moment.  Here is a brief example of what the difference
would be between linear FM modulations and delta FM modulation:
 
    Original	Linear		Delta		Change in
    Signal	Frequency	Frequency	Signal
    Amplitude	Modulation	Modulation	Amplitude
 
      0 dB	10.0 MHz	10.0 MHz	  0 dB
      5 dB	10.1 MHz	10.1 MHz	  5 dB
     10 dB	10.2 MHz	10.1 MHz	  5 dB
     10 dB	10.2 MHz	10.0 MHz	  0 dB
     15 dB	10.3 MHz	10.1 MHz	  5 dB
     20 dB	10.5 MHz	10.2 MHz	  5 dB
    -10 dB	 9.0 MHz	 8.5 MHz	-30 dB
 
What the two systems have in common is that you designate a frequency,
10.0 MHz in this case, to be your carrier.
 
Now here is where the difference is.  An absolute level of 0 dB
corresponds to the carrier frequency in linear FM.  That means that
every frequency has one,  and only one, possible signal level
associated with it.  0 dB is 10.0 MHz, 5 dB is 10.1 MHz, and so on.
The formula, in this case, would be
 
    Frequency = Carrier  + (Amplitude * a constant)
    Frequency = 10.0 MHz + (Amplitude * 0.1 MHz)
 
One of the implications of this is that linear FM is not context
sensitive.  You can take any small randomly selected portion of
the signal and deduce the original waveform from it.
 
Delta FM, on the other hand, expresses a *change* in level, rather
than an absolute level.  Thus, the carrier frequency, 10.0 MHz in
this case, corresponds to a change in level of 0 dB.  If the previous
level is the same as the current level, no matter what it is, the
transmitter will send a frequency of 10.0 MHz.  Thus you can get
a 10.0 MHz signal for *any* level, so long as it's the same as the
previous one.  For example, if the previous signal level were 0 dB
and the current one is 5 dB the transmitter would transmit a frequency
of 10.1 MHz because 5 dB - 0 dB = 5 dB.  If the previous signal
were 10 dB and the new one were 15 dB, the transmitter would also
transmit a frequency of 10.1 MHz because 15 dB - 10 dB = 5 dB.  The
formula here is:
 
    Freq. = Carrier  + ((Current Amplitude - Previous Amp.) * Constant)
    Freq. = 10.0 MHz + ((Current Amplitude - Previous Amp.) * 0.1 Mhz)
 
Here, any particular frequency could correspond to any particular
amplitude level, depending on what the amplitude level before that
was.  You can't decode the original signal from any small random
portion of the transmitter signal because you have to find out what
the current frequency at the the beginning of your segment was.
 
In short, our current system of modulation is linear FM, not delta
FM.  I've explained why, for the curious.  I hope this clears up
this mess, and that the explanation wasn't overly technical.
 
cjs
--
                        | "It is actually a feature of UUCP that the map of
[log in to unmask]         | all systems in the network is not known anywhere."
[log in to unmask] |    --Berkeley Mail Reference Manual (Kurt Schoens)

ATOM RSS1 RSS2