) writes:
Michael Black wrote:

I glanced at it and maybe missed something, but DSB is AM. And
he certainly says it at the outset, and when he's talking about the
components he's talking about 2 sidebands and a carrier.

Now, "DSB" often has fallen into the meaning of "DSB with no carrier",
but technically one should specifically define that there is no carrier.
snip
Michael

Back in 1972 when I took my FFC 2nd and 1st class exams DSB was defined
as the sidebands with a supressed carrier. A signal with both sidebands
and the
carrier was simply AM with a BW disgnator. .Now that diffintion may
have slipped
over the years, but from my perspective AM means both sidebands, with a
carier
DSB means both sidebands without the carrier, and ISB means two
different
sidebands with no carrier. I only have received the later, ISB, a very
few times
mainly on ancient STL links.

It might be useful to check out what the ITU says these days about
"AM", both sidebands with carrier", and for this conversation, "DSB"
being both sidebands without the carrier.

Terry

Are you arguing semantics, or understanding?

The post I replied to was almost outraged by that PDF's useage of "DSB".
I couldn't figure out whether he was just fussing over words (and thus
why was he so outraged?), or really does believe that DSB is not AM.

Because people have become sloppy about the words, some of all
these conversations about "better AM detectors" is limited. Because
people are searching for something that really isn't all that different
from what's already available. That PDF talks in terms of how
synchronous detectors get too much hype, yet the author turns around
and uses everything a "synchronous detector" has except the actual
synchronization. But the synchronization isn't actually what provides
the potentially improved reception, it's just a means of compensating
for some side effects.

I never got around to replying, but a few months ago someone started
a thread here where he stated something like "So I gather the carrier
is more likely to fade when selective fading is happening". I haven't
a clue whether the carrier is more likely to fade than the sidebands, but
once the carrier fades in relationship to the sidebands you're going to
start having reception problems, and once the carrier completely fades
you won't be able to recover the modulation. The carrier is the key
part to demodulation. But a more universal understanding of "amplitude
modulation" would show right away that you can't demodulate a DSB
signal unless a carrier is sent along, or generated locally at the receiver
end, and selective fading can mean that a DSBc signal sent from the
transmitter may be a DSBsc (Double SIdeband suppressed carrier) by the
time it reaches the receiver.

So in this sort of talk, you'd better start being specific about what
you are talking about. Since DSB (with or without a carrier) and
SSB (with or without a carrier) are "AM", then you really need to
stop using "AM" to only mean DSBc.

Hence DSB in the PDF is more descriptive than AM. Is he confused?
I don't think so. In his opening paragraph he says "Note: DSB (Double
Sideband full-carrier) and SSB (single sideband suppressed carrier) are
both amplitude modulation". He defines the term as he is about to
use them, so there is no confusion. He needs to use the DSB rather
than a more generic "AM" because he is very much thinking in terms
of two sidebands (even if he turns around and removes one). The
fact that there are two sidebands rather than one may be more significant
than whether or not there is a carrier.

Since he defined his terms to begin with, any subsequent useage of
"DSB" is taken care of. But, again, even if that was not the case,
his useage is fine, because whether or not a carrier is sent is
irrelevant to his discussion. It's easy to get a locally generated
"carrier", and if it's just one sideband it's done all the time, with
a bit of mistuning. But with two sidebands, it's far harder. Hence
you can either determine where the locally generated carrier needs to
be from the the redundant sidebands, or strip off one sideband so it
becomes SSB and placing the carrier becomes much easier.

Maybe he should have gone with DSBc to show that he is talking about
a DSB signal with carrier, but that is hardly a confusion of AM and DSB.

As for common useage of so many of these terms, nobody had to specify
how many sidebands and whether a carrier went with it until they
started to use a subset of that stuff. Look in early articles about
SSB and it was pretty common for them to be most specific, ie SSBsc (SSB
suppressed carrier). It's only later that it simply became SSB. Nobody
really thought of sending DSB without a carrier until SSB came along,
and there too it was not uncommon to see it referred to as DSBsc.

Michael

Michael Black wrote:
) writes:
Michael Black wrote:

I glanced at it and maybe missed something, but DSB is AM. And
he certainly says it at the outset, and when he's talking about the
components he's talking about 2 sidebands and a carrier.

Now, "DSB" often has fallen into the meaning of "DSB with no carrier",
but technically one should specifically define that there is no carrier.
snip
Michael

Back in 1972 when I took my FFC 2nd and 1st class exams DSB was defined
as the sidebands with a supressed carrier. A signal with both sidebands
and the
carrier was simply AM with a BW disgnator. .Now that diffintion may
have slipped
over the years, but from my perspective AM means both sidebands, with a
carier
DSB means both sidebands without the carrier, and ISB means two
different
sidebands with no carrier. I only have received the later, ISB, a very
few times
mainly on ancient STL links.

It might be useful to check out what the ITU says these days about
"AM", both sidebands with carrier", and for this conversation, "DSB"
being both sidebands without the carrier.

Terry

Are you arguing semantics, or understanding?

It is not sematics, but generally agreed upon definitions.

The post I replied to was almost outraged by that PDF's useage of "DSB".
I couldn't figure out whether he was just fussing over words (and thus
why was he so outraged?), or really does believe that DSB is not AM.

Because people have become sloppy about the words, some of all
these conversations about "better AM detectors" is limited. Because
people are searching for something that really isn't all that different
from what's already available. That PDF talks in terms of how
synchronous detectors get too much hype, yet the author turns around
and uses everything a "synchronous detector" has except the actual
synchronization. But the synchronization isn't actually what provides
the potentially improved reception, it's just a means of compensating
for some side effects.

I never got around to replying, but a few months ago someone started
a thread here where he stated something like "So I gather the carrier
is more likely to fade when selective fading is happening". I haven't
a clue whether the carrier is more likely to fade than the sidebands, but
once the carrier fades in relationship to the sidebands you're going to
start having reception problems, and once the carrier completely fades
you won't be able to recover the modulation. The carrier is the key
part to demodulation. But a more universal understanding of "amplitude
modulation" would show right away that you can't demodulate a DSB
signal unless a carrier is sent along, or generated locally at the receiver
end, and selective fading can mean that a DSBc signal sent from the
transmitter may be a DSBsc (Double SIdeband suppressed carrier) by the
time it reaches the receiver.

So in this sort of talk, you'd better start being specific about what
you are talking about. Since DSB (with or without a carrier) and
SSB (with or without a carrier) are "AM", then you really need to
stop using "AM" to only mean DSBc.

Hence DSB in the PDF is more descriptive than AM. Is he confused?
I don't think so. In his opening paragraph he says "Note: DSB (Double
Sideband full-carrier) and SSB (single sideband suppressed carrier) are
both amplitude modulation". He defines the term as he is about to
use them, so there is no confusion. He needs to use the DSB rather
than a more generic "AM" because he is very much thinking in terms
of two sidebands (even if he turns around and removes one). The
fact that there are two sidebands rather than one may be more significant
than whether or not there is a carrier.

Since he defined his terms to begin with, any subsequent useage of
"DSB" is taken care of. But, again, even if that was not the case,
his useage is fine, because whether or not a carrier is sent is
irrelevant to his discussion. It's easy to get a locally generated
"carrier", and if it's just one sideband it's done all the time, with
a bit of mistuning. But with two sidebands, it's far harder. Hence
you can either determine where the locally generated carrier needs to
be from the the redundant sidebands, or strip off one sideband so it
becomes SSB and placing the carrier becomes much easier.

Maybe he should have gone with DSBc to show that he is talking about
a DSB signal with carrier, but that is hardly a confusion of AM and DSB.

As for common useage of so many of these terms, nobody had to specify
how many sidebands and whether a carrier went with it until they
started to use a subset of that stuff. Look in early articles about
SSB and it was pretty common for them to be most specific, ie SSBsc (SSB
suppressed carrier). It's only later that it simply became SSB. Nobody
really thought of sending DSB without a carrier until SSB came along,
and there too it was not uncommon to see it referred to as DSBsc.

I just don't buy the argument that nobody thought about DSB. DSB is
naturally generated from a mixer if the audio signal is zero mean.
[Again, it is best to talk about modulation schemes by discussion
modulators.] Remember, AM exists because the demod is cheaper, not
because it is any easier to generate that say DSB. AM and DSB use the
same hardware.

Michael

To get to the meat of the problem, the only utility in any of these
demod schemes is if you can narrow band the signal. That is, just
because you can build a demodulator that can use both sidebands without
need of the carrier, it doesn't mean the quality of the signal will be
any better. Wider bandwidth signals are more prone to atmospheric
effects, i.e. fading. Thus if you are going to do synchronous
detection, you need to receive one side band, period.

In article . com,
wrote:
I just don't buy the argument that nobody thought about DSB. DSB is
naturally generated from a mixer if the audio signal is zero mean.
[Again, it is best to talk about modulation schemes by discussion
modulators.] Remember, AM exists because the demod is cheaper, not
because it is any easier to generate that say DSB. AM and DSB use the
same hardware.

But there's something where a DSB local carrier has to be in phase with
the signal, otherwise there some sort of canceling effect from the two
sidebands. SSB doesn't have the problem, it just causes a pitch shift.

The only DSB I know of, (that isn't just treated as a cheap form of SSB,
where really only one sideband is used) is the L-R subcarrier in FM stereo
where they double the pilot tone to get an in phase carrier.

Mark Zenier
Googleproofaddress(account:mzenier provider:eskimo domain:com)

Mark Zenier wrote:
In article . com,
wrote:
I just don't buy the argument that nobody thought about DSB. DSB is
naturally generated from a mixer if the audio signal is zero mean.
[Again, it is best to talk about modulation schemes by discussion
modulators.] Remember, AM exists because the demod is cheaper, not
because it is any easier to generate that say DSB. AM and DSB use the
same hardware.

But there's something where a DSB local carrier has to be in phase with
the signal, otherwise there some sort of canceling effect from the two
sidebands. SSB doesn't have the problem, it just causes a pitch shift.

I've done DSB generation in a baseband signal for a telemetry
application, similar to the FM stereo demod, i.e a pilot was provided.
However, say you had no pilot and you were simply tuning by hand.
Assume a simple mixer. If you were off a bit, the sound would be pretty
ugly, i.e. 1000hz would be a combination of 999hz and 1001Hz if you
were off by 1Hz in the local carrier. However, I could see this making
tuning very easy since if you were off a little, it would be clear that
the local carrier needed adjusting.


The only DSB I know of, (that isn't just treated as a cheap form of SSB,
where really only one sideband is used) is the L-R subcarrier in FM stereo
where they double the pilot tone to get an in phase carrier.

Mark Zenier
Googleproofaddress(account:mzenier provider:eskimo domain:com)