Frequency offset compensation for 802.15.4 (ZigBee-MSK)

Eric Jacobsen <eric.jacobsen@ieee.org> wrote:

>On Wed, 7 Oct 2015 23:34:02 +0000 (UTC), spope33@speedymail.org (Steve
>Pope) wrote:

>>Eric Jacobsen <eric.jacobsen@ieee.org> wrote:

>>> Filtering does increase the PAPR with PSK, but it is not the
>>> reason AM is there, and it doesn't have that much to do with
>>> how long the pulse shape is.   It has AM even with a pulse
>>> shape one symbol long.

>>You've said this last sentence a couple times too, and I can't
>>really agree with it.  PSK with a rectangular pulse shape one 
>>symbol long has constant energy per baud.

>>A.M. is usually considered to be amplitude modulation of the envelope of 
>>the carrier, and I do not see how PSK with the above pulse shape 
>>has any A.M.

>>Nor do I see where the PAPR of this signal is different from
>>that of an unmodulated carrier.  Unless there's further filtering.

>Do you think that PSK is only PSK in the strict condition that it only
>has rectangular pulses?    

Not at all, upthread I myself discussed PSK in which the pulse shape is
an impulse.

But it's your immediate statment here that PSK "... has AM even with
a pulse shape one symbol long".  Given a rectangular pulse
shape one symbol long, I do not see how this can possibly be the case.  
It has no more AM than an unmodulated carrier, that is to say, zero.

>Is it still PSK if there is a matched
>filter, say an RRC, on both ends of the link?  

Again, a terminology question, and we seem to be having some 
difficulty with those.  But I did say in my last post that
"PSK with a little AM is still PSK" in most normal contexts.

>>>>>What counts in this discussion is the waveform of the signal being 
>>>>>transmitted.  Processing within the receiver doesn't count.

>>>>In a PSK system half of the pulse matched-filtering is typically done
>>>>in the receiver, so what's done in the receiver shouldn't be ignored.
>>>>It's part of the system design.

>>I'm not saying to ignore receiver processing on a global basis;
>>but when answering questions of the form, "What type(s) of modulation
>>does the transmitted waveform exhibit" (which is I believe the
>>topic at hand), what's done in the receiver is not relevant.

>Whether there's AM on the constellation, i.e., the modulation where
>the information is contained, is only discernible inside the
>demodulator after the matched filter.    

Correct 

> So what's done inside the demodulator is, in fact, relevant.

It's relevant to many questions, but not to the question of what
type(s) of modulation is(are) being transmitted.  

Steve

On Thu, 8 Oct 2015 02:27:29 +0000 (UTC), spope33@speedymail.org (Steve
Pope) wrote:

>Eric Jacobsen <eric.jacobsen@ieee.org> wrote:
>
>>On Wed, 7 Oct 2015 23:34:02 +0000 (UTC), spope33@speedymail.org (Steve
>>Pope) wrote:
>
>>>Eric Jacobsen <eric.jacobsen@ieee.org> wrote:
>
>>>> Filtering does increase the PAPR with PSK, but it is not the
>>>> reason AM is there, and it doesn't have that much to do with
>>>> how long the pulse shape is.   It has AM even with a pulse
>>>> shape one symbol long.
>
>>>You've said this last sentence a couple times too, and I can't
>>>really agree with it.  PSK with a rectangular pulse shape one 
>>>symbol long has constant energy per baud.
>
>>>A.M. is usually considered to be amplitude modulation of the envelope of 
>>>the carrier, and I do not see how PSK with the above pulse shape 
>>>has any A.M.
>
>>>Nor do I see where the PAPR of this signal is different from
>>>that of an unmodulated carrier.  Unless there's further filtering.
>
>>Do you think that PSK is only PSK in the strict condition that it only
>>has rectangular pulses?    
>
>Not at all, upthread I myself discussed PSK in which the pulse shape is
>an impulse.
>
>But it's your immediate statment here that PSK "... has AM even with
>a pulse shape one symbol long".  Given a rectangular pulse
>shape one symbol long, I do not see how this can possibly be the case.  
>It has no more AM than an unmodulated carrier, that is to say, zero.

So there is one pulse shape, where only in theoretical circumstances
with infinite bandwidth, the signal can transition 180 degrees in zero
time and result in no AM.   Is that a definining case?    Because in
any other case, and in all practical cases, there will be AM on a PSK
signal any time the signal make a 180-degree transition.  

There are many other pulse shapes that are only one symbol long that
produce a lot of AM, including the one that produces zero AM with
OQPSK, i.e., the half-sine symbol pulse.

>>Is it still PSK if there is a matched
>>filter, say an RRC, on both ends of the link?  
>
>Again, a terminology question, and we seem to be having some 
>difficulty with those.  But I did say in my last post that
>"PSK with a little AM is still PSK" in most normal contexts.

How much is a little?  RRC filters at about 12.5% EBW increase PAPR
approaching 3dB compared to 100% EBW.    Is it still PSK?   If a
little is okay, why isn't more okay?    What is the criterion for how
much is too much?  What definition of PSK says there can't be PAPR?
Why do all the vendors who make modems with filtering call the
modulation PSK? You've done much to support your position, while I've
been offering an awful lot of evidence to the contrary.

Specifically, the issue is your stated position that:

>I will counter by saying that, in the general case, filtering 
>a phase-modulated signal will add amplitude modulation, thereby moving it 
>outside of the definition of being purely phase modulated.

Try to support that with something other than just repeated
assertions.   I've shown multiple times why this is not true, and many
of the points are very clear but go ignored, like the evidence
presented in a common eye diagram.   Instead you return to the
infinite-bandwidth rectangular pulse case, which is not practical in
any sense.

I suspect you have an alternate agenda here, because you seem to be
ignoring the evidence and I know you're smarter than that.

>>>>>>What counts in this discussion is the waveform of the signal being 
>>>>>>transmitted.  Processing within the receiver doesn't count.
>
>>>>>In a PSK system half of the pulse matched-filtering is typically done
>>>>>in the receiver, so what's done in the receiver shouldn't be ignored.
>>>>>It's part of the system design.
>
>>>I'm not saying to ignore receiver processing on a global basis;
>>>but when answering questions of the form, "What type(s) of modulation
>>>does the transmitted waveform exhibit" (which is I believe the
>>>topic at hand), what's done in the receiver is not relevant.
>
>>Whether there's AM on the constellation, i.e., the modulation where
>>the information is contained, is only discernible inside the
>>demodulator after the matched filter.    
>
>Correct 
>
>> So what's done inside the demodulator is, in fact, relevant.
>
>It's relevant to many questions, but not to the question of what
>type(s) of modulation is(are) being transmitted.  

It's specifically relevant to whether the AM added by pulse filtering
in PSK affects the modulation.  i.e., on-topic.


Eric Jacobsen
Anchor Hill Communications
http://www.anchorhill.com

Eric Jacobsen <eric.jacobsen@ieee.org> wrote:

>On Thu, 8 Oct 2015 02:27:29 +0000 (UTC), spope33@speedymail.org (Steve

>>>>> Filtering does increase the PAPR with PSK, but it is not the
>>>>> reason AM is there, and it doesn't have that much to do with
>>>>> how long the pulse shape is.   It has AM even with a pulse
>>>>> shape one symbol long.

>>>>You've said this last sentence a couple times too, and I can't
>>>>really agree with it.  PSK with a rectangular pulse shape one 
>>>>symbol long has constant energy per baud.
>>
>>>>A.M. is usually considered to be amplitude modulation of the envelope of 
>>>>the carrier, and I do not see how PSK with the above pulse shape 
>>>>has any A.M.

>>>Do you think that PSK is only PSK in the strict condition that it only
>>>has rectangular pulses?    

>>Not at all, upthread I myself discussed PSK in which the pulse shape is
>>an impulse.

>>But it's your immediate statment here that PSK "... has AM even with
>>a pulse shape one symbol long".  Given a rectangular pulse
>>shape one symbol long, I do not see how this can possibly be the case.  
>>It has no more AM than an unmodulated carrier, that is to say, zero.

>So there is one pulse shape, where only in theoretical circumstances
>with infinite bandwidth, the signal can transition 180 degrees in zero
>time and result in no AM.   Is that a definining case?    

Yes, it's a defining case in the sense that if you take a carrier,
and you digitally phase-modulate it with a stream of bits (or dibits, 
tribits etc.), and you do nothing else, this is what you get.  

>Specifically, the issue is your stated position that:

>>I will counter by saying that, in the general case, filtering 
>>a phase-modulated signal will add amplitude modulation, thereby moving it 
>>outside of the definition of being purely phase modulated.

>Try to support that with something other than just repeated
>assertions.   I've shown multiple times why this is not true

If you feel you have shown this, good for you.

> Instead you return to the infinite-bandwidth rectangular pulse case, 
> which is not practical in any sense.

Yes, in reality, all PSK signals are filtered beyond the implicit
filtering of a rectangular pulse shape, and therefore exhibit A.M.

(It's also true that in reality, all MSK signals are similarly
filtered, beyond the implicit filtering in a half-sine pulse shape,
and therefore deviate from the definition of MSK.)

>I suspect you have an alternate agenda here, because you seem to be
>ignoring the evidence and I know you're smarter than that.

No agenda other than discussing the issue factually.

Steve

On Thu, 8 Oct 2015 13:22:18 +0000 (UTC), spope33@speedymail.org (Steve
Pope) wrote:

>Eric Jacobsen <eric.jacobsen@ieee.org> wrote:
>
>>On Thu, 8 Oct 2015 02:27:29 +0000 (UTC), spope33@speedymail.org (Steve
>
>>>>>> Filtering does increase the PAPR with PSK, but it is not the
>>>>>> reason AM is there, and it doesn't have that much to do with
>>>>>> how long the pulse shape is.   It has AM even with a pulse
>>>>>> shape one symbol long.
>
>>>>>You've said this last sentence a couple times too, and I can't
>>>>>really agree with it.  PSK with a rectangular pulse shape one 
>>>>>symbol long has constant energy per baud.
>>>
>>>>>A.M. is usually considered to be amplitude modulation of the envelope of 
>>>>>the carrier, and I do not see how PSK with the above pulse shape 
>>>>>has any A.M.
>
>>>>Do you think that PSK is only PSK in the strict condition that it only
>>>>has rectangular pulses?    
>
>>>Not at all, upthread I myself discussed PSK in which the pulse shape is
>>>an impulse.
>
>>>But it's your immediate statment here that PSK "... has AM even with
>>>a pulse shape one symbol long".  Given a rectangular pulse
>>>shape one symbol long, I do not see how this can possibly be the case.  
>>>It has no more AM than an unmodulated carrier, that is to say, zero.
>
>>So there is one pulse shape, where only in theoretical circumstances
>>with infinite bandwidth, the signal can transition 180 degrees in zero
>>time and result in no AM.   Is that a definining case?    
>
>Yes, it's a defining case in the sense that if you take a carrier,
>and you digitally phase-modulate it with a stream of bits (or dibits, 
>tribits etc.), and you do nothing else, this is what you get.  
>
>>Specifically, the issue is your stated position that:
>
>>>I will counter by saying that, in the general case, filtering 
>>>a phase-modulated signal will add amplitude modulation, thereby moving it 
>>>outside of the definition of being purely phase modulated.
>
>>Try to support that with something other than just repeated
>>assertions.   I've shown multiple times why this is not true
>
>If you feel you have shown this, good for you.
>
>> Instead you return to the infinite-bandwidth rectangular pulse case, 
>> which is not practical in any sense.
>
>Yes, in reality, all PSK signals are filtered beyond the implicit
>filtering of a rectangular pulse shape, and therefore exhibit A.M.
>
>(It's also true that in reality, all MSK signals are similarly
>filtered, beyond the implicit filtering in a half-sine pulse shape,
>and therefore deviate from the definition of MSK.)

One can make a practical MSK implementation much, much closer to
"ideal" MSK than one can make a rectangular-pulse PSK system with
infinite bandwidth.

>>I suspect you have an alternate agenda here, because you seem to be
>>ignoring the evidence and I know you're smarter than that.
>
>No agenda other than discussing the issue factually.

People often make their own definitions of "fact".   Some stand up to
evidence better than others.


Eric Jacobsen
Anchor Hill Communications
http://www.anchorhill.com

Eric Jacobsen <eric.jacobsen@ieee.org> wrote:

>On Thu, 8 Oct 2015 13:22:18 +0000 (UTC), spope33@speedymail.org (Steve

>>No agenda other than discussing the issue factually.

>People often make their own definitions of "fact".   Some stand up to
>evidence better than others.

One fact is that you haven't been able to point out anything
factually incorrect that I have stated in this thread.

You've engaged in many oblique arguments, along with challenges to
choices of terminlogy that are so fine-tuned as to be of little
practical import.  

Why you are doing this, I have no idea.

Steve

On Thu, 8 Oct 2015 20:05:39 +0000 (UTC), spope33@speedymail.org (Steve
Pope) wrote:

>Eric Jacobsen <eric.jacobsen@ieee.org> wrote:
>
>>On Thu, 8 Oct 2015 13:22:18 +0000 (UTC), spope33@speedymail.org (Steve
>
>>>No agenda other than discussing the issue factually.
>
>>People often make their own definitions of "fact".   Some stand up to
>>evidence better than others.
>
>One fact is that you haven't been able to point out anything
>factually incorrect that I have stated in this thread.

Actually, that's incorrect, too.  I did, actually, several times.
But I can re-iterate:

First:
>Maybe the disconnect here is that you are viewing OQPSK as
>something that evolved from QPSK.  It did not.  Designers
>who were already going to use MSK anyway chose to use the OQPSK 
>terminology in some standards submissions, for what I believe
>were non-technical reasons.  To a first order that is all
>OQPSK is.

Second:
>While these aren't explicit statements about origin, they seem
>to downplay that MSK in fact originated from FM; and that OQPSK is 
>in fact terminology that came along later.

The above were both factually incorrect, but you did later say you
actually didn't know after I pointed out the history:

>>You're explicitly stating this as fact again without any evidence.  I
>>think the MSK terminology came along much later.
>
>This could be; I don't know for certain.  I first encountered the MSK
>terminology in one of Oldenwater's papers, which was from the 70's, that 
>I personally ran across in the 80's.  It did not seem like he was 
>introducing a new term.

PSK pre-dated that by a couple of decades or so.  I cited sources for
the history, plus my own experience in the industry replacing early
equipment.

Third:
>>If you apply any kind of filtering to MSK, it's not MSK any more.
>
>True.  The exact same could be said about PSK.   

You later retreated by saying you meant something else.   

>I said it "could be said" that filtering changes PSK into something other 
>than PSK.  Saying "is could be said X" is not the same as saying "X".

I don't think I've been the one being oblique.   Or maybe it's obtuse,
I dunno.  

Fourth:
>I will counter by saying that, in the general case, filtering 
>a phase-modulated signal will add amplitude modulation, thereby moving it 
>outside of the definition of being purely phase modulated.

You later tried to claim you said something else, specifically "can"
instead of "will", and that speaking of "the general case" meant it
didn't strictly apply.   Regardless, the stated point that a filter
that adds AM to PSK will move it "outside of the definition of being
purely phase modulated" still remains, and is still incorrect. 

Clearly PSK systems rely on matched filters for spectral containment
and performance, and those filters typically add more and more AM as
the spectral containment tightens, which is routinely done in
practice.   This can be seen on any typical eye diagram, yet they are
most certainly not "outside the definition of being purely phase
modulated".   The information is only in the phase changes, and since
those phase changes can include 180-degree transitions (and other
transitions that affect the envelope), there is naturally a
significant amount of amplitude deviation in the PSK signal to
accomodate those phase changes.   

So PSK does not get moved "outside of the definition of being purely
phase modulated" by adding filtering, even if the filtering increases
the PAPR.

Fifth:
>One fact is that you haven't been able to point out anything
>factually incorrect that I have stated in this thread.

Also not true, as noted above.

>You've engaged in many oblique arguments, along with challenges to
>choices of terminlogy that are so fine-tuned as to be of little
>practical import.  

I disagree.   It seems to me you're the one that's been oblique (or
obtuse, however you want to look at it), for example by ignoring the
cited eye diagrams and retreating to the infinite bandwidth,
rectangular-pulse PSK definition, apparently to defend your position
that PSK has no AM or it isn't PSK.   

>Why you are doing this, I have no idea.

I'm enjoying the dialogue, and in this area I don't want to leave the
errors unchallenged.   I've no idea what your motivation is.


Eric Jacobsen
Anchor Hill Communications
http://www.anchorhill.com

   
> 
> So PSK does not get moved "outside of the definition of being purely
> phase modulated" by adding filtering, even if the filtering increases
> the PAPR.
>

As I see it, if you plot the signal ( including the __trajectory__ between the data points) on the IQ plane and the result is anything but a circle, then the signal has some AM modulation.

Mark

On Fri, 9 Oct 2015 05:51:27 -0700 (PDT), makolber@yahoo.com wrote:

>   
>> 
>> So PSK does not get moved "outside of the definition of being purely
>> phase modulated" by adding filtering, even if the filtering increases
>> the PAPR.
>>
>
>As I see it, if you plot the signal ( including the __trajectory__ between the data points) on the IQ plane and the result is anything but a circle, then the signal has some AM modulation.

I think a lot of people would agree with that (myself included), and
PSK, even with infinite-bandwidth rectangular pulses, has AM by that
definition.


Eric Jacobsen
Anchor Hill Communications
http://www.anchorhill.com