16QAM and 16PSK

Please  can anyone give me a pointer to a material that does an in
depth comparison of the 16QAM and 16PSK modulation in terms of
different properties and their SNR vs BER comparison. Any comment
about this modulation comparison would be helpful couldn't turn up
anything using google and even IEEE.

Thanks

On 01/07/2011 04:37 PM, Deamon wrote:
> Please  can anyone give me a pointer to a material that does an in
> depth comparison of the 16QAM and 16PSK modulation in terms of
> different properties and their SNR vs BER comparison. Any comment
> about this modulation comparison would be helpful couldn't turn up
> anything using google and even IEEE.

Not in depth, but:

(anything)PSK gives you a constant envelope to transmit with, which 
reduces the complexity of your transmitter's circuitry.

Now, plot out a circle on a piece of paper, and draw 16 evenly spaced 
dots on that circle -- that's the constellation for 16PSK.  Then 
inscribe a square within that circle, and populate the square with 16 
evenly spaced dots.  That's the constellation for 16QAM with the same 
maximum transmit power as your 16PSK.  Which dots have more distance 
between them?  If you think that the same maximum transmit power is to 
stringent a comparison, then figure out how to scale the square for the 
same _average_ power -- now compare the distance between the dots.

16QAM -- better noise performance (at least if it's additive Gaussian).

16PSK -- easier transmitter design.

I've only personally encountered PSK up to 8, but I don't do slews of 
comms design.

-- 

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" was written for you.
See details at http://www.wescottdesign.com/actfes/actfes.html

On Jan 7, 6:37&#4294967295;pm, Deamon <persistence...@gmail.com> wrote:
> Please &#4294967295;can anyone give me a pointer to a material that does an in
> depth comparison of the 16QAM and 16PSK modulation in terms of
> different properties and their SNR vs BER comparison. Any comment
> about this modulation comparison would be helpful couldn't turn up
> anything using google and even IEEE.
>
> Thanks

In addition to following Tim Wescott's excellent suggestions
which will *really* help you a lot, you might want to look at a
document titled Signal Space Concepts that is available at

<http://courses.engr.illinois.edu/ece361/spring11/OldLectures.html>

--Dilip Sarwate

Tim Wescott wrote:


> (anything)PSK gives you a constant envelope to transmit with, which 
> reduces the complexity of your transmitter's circuitry.

Except for bandlimited PSK does not have a constant envelope; in the 
other words, no advantage.

> I've only personally encountered PSK up to 8,

QAM clearly outperforms PSK at n >= 8. Consequently, there is no point 
in using more then 8-PSK.


Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com

>
>
>Tim Wescott wrote:
>
>
>> (anything)PSK gives you a constant envelope to transmit with, which 
>> reduces the complexity of your transmitter's circuitry.
>
>Except for bandlimited PSK does not have a constant envelope; in the 
>other words, no advantage.
>
>> I've only personally encountered PSK up to 8,
>
>QAM clearly outperforms PSK at n >= 8. Consequently, there is no point 
>in using more then 8-PSK.

You are just looking at this from a signal processing point of view. As Tim
said, bring the transmitter into the analysis and things may look
different. A constant power modulation allows options in transmitter design
that can offer reduced cost (which Tim said) and higher efficiency (which
he should have). Its not a clear choice until the constellation gets
bigger, and PSK starts to look really bad.

Steve

steveu wrote:

>>
>>Tim Wescott wrote:
>>
>>
>>
>>>(anything)PSK gives you a constant envelope to transmit with, which 
>>>reduces the complexity of your transmitter's circuitry.
>>
>>Except for bandlimited PSK does not have a constant envelope; in the 
>>other words, no advantage.
>>
>>
>>>I've only personally encountered PSK up to 8,
>>
>>QAM clearly outperforms PSK at n >= 8. Consequently, there is no point 
>>in using more then 8-PSK.
> 
> 
> You are just looking at this from a signal processing point of view. As Tim
> said, bring the transmitter into the analysis and things may look
> different.

Like what?

> A constant power modulation allows options in transmitter design

However, bandlimited PSK is not a constant power modulation.

> that can offer reduced cost (which Tim said) and higher efficiency (which
> he should have).

How?

> Its not a clear choice until the constellation gets
> bigger, and PSK starts to look really bad.

Example?


VLV

On Jan 8, 2:37&#4294967295;am, Deamon <persistence...@gmail.com> wrote:
> Please &#4294967295;can anyone give me a pointer to a material that does an in
> depth comparison of the 16QAM and 16PSK modulation in terms of
> different properties and their SNR vs BER comparison. Any comment
> about this modulation comparison would be helpful couldn't turn up
> anything using google and even IEEE.
>
> Thanks

I found the following book a good source for comparison of different
modulation schemes:
"Digital Modulation Techniques" by Fuqin XIONG

The author compares QAM and PSK to show that QAM is superior to MPSK.
Specifically for M=16 the difference is about 4.2dB (quoted from the
book - "all rights reserved").
A simple explanation to this difference in favor of QAM modulation is
that QAm uses both dimensions while PSK only uses the phase.

Although MPSK only uses the phase to place the constellation it is not
a constant envelope modulation. The signal still "needs" to move from
one point to the other. Some methods are known to generate QPSK or
rather OQPSK to constant or quasi-constant envelope but in the cost of
increasing the spectral mask.

Moti

On 01/07/2011 07:37 PM, Deamon wrote:
> Please  can anyone give me a pointer to a material that does an in
> depth comparison of the 16QAM and 16PSK modulation in terms of
> different properties and their SNR vs BER comparison. Any comment
> about this modulation comparison would be helpful couldn't turn up
> anything using google and even IEEE.
>
> Thanks

One must ask themselves, "On what basis am I comparing?" In other words,
what is important to me?

Two common bases are: bandwidth efficiency and power efficiency. This
handout from Brian Hughes (part of my information theory class at
NCSU) explores this. Especially, see Figure 2.2.

   http://www.digitalsignallabs.com/modulation.pdf

--Randy

-- 
Randy Yates                      % "My Shangri-la has gone away, fading like
Digital Signal Labs              %  the Beatles on 'Hey Jude'"
yates@digitalsignallabs.com      %
http://www.digitalsignallabs.com % 'Shangri-La', *A New World Record*, ELO