[Laser] Polarization modulation

[C. Turner]

Hi Tom,

I was mentioning the use of orthogonal modulation mainly because it 
presents the best-case scenario in terms of S/N.  Although the analogy 
isn't perfect, it could be likened to any N-ary phase modulation in 
which case the best S/N is obtained with the largest phase change (e.g. 
BPSK trumps QPSK, etc.)

Were S/N not a main concern, one could modulate with an arbitrary angle 
change - however that might be done - possibly with a crystal...  Unlike 
with RF where one could resolve the phase implicitly, determining the 
polarity with, say, light would be a bit trickier - unless there's some 
clever way to do this of which I'm not aware.

One could determine the angle by observing the amplitude ratio between 
the H and V detectors, but this has a level of uncertainty to it - but 
it is probably "good enough" when a good S/N was present.

Were one to use optical filters, determining polarity by measuring angle 
at which the signal peaks is less-accurate than finding its null and 
then offsetting by 90 degrees - but doing it that way (unless you had an 
array of detectors and polarizers, a bunch of Brewster's windows or 
something like that) would tricky and imply higher losses.

* * *

One interesting scheme would be to use orthogonal modulation (e.g. 
separate H and V) from the same aperture and detect them independently 
as H and V.  In theory, one could simultaneously send two things at once 
in several ways:

- The obvious is H for one "channel" and V for the other channel.  If 
one listened on a polarity-insensitive system one would hear both 
channels mixed together.

- Another would be to modulate both H and V together for one channel and 
then differentially modulate the two for the other - but keeping the 
TOTAL power in the H and V channels combined constant.  (Does that 
remind you of something else?)  In this way, the differential H/V 
channel would be "invisible" on a receiver that was polarity insensitive 
(and assuming no polarization favoring occurring along the optical path 
and/or in the lenses/detectors used, of course.)  There are obvious 
limitations and complications - the most apparent being that you'd never 
be able to allow the power in the "H+V" channel to go to zero, etc. and 
the need to compensate for amplitude shifts related to H+V power in the 
differential channel (something that could be compensated in a 
variable-gain stage slaved from the H+V combined signal) but it's an 
interesting thought, anyway.

73,

Clint
KA7OEI



>
> ------------------------------
>
> Message: 2
> Date: Thu, 10 Mar 2011 19:48:26 -0500
> From: Tom Becker<GTBecker at RighTime.com>
> Subject: Re: [Laser] Polarization modulation (Was: High-power
> 	polarized	light?)
> To: Free Space LASER Communications<laser at mailman.qth.net>
> Message-ID:<4D79715A.1030408 at RighTime.com>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>
> Polarization need not be binary; it can be any angle, can it not?
>
> Instead of considering a polarized beam as two fixed-angle differential
> channels, might there be an advantage of phase modulation, using
> polarization as the medium?  How to convert phase to polarity is an
> interesting question but, once answered, why not use the entire rotation
> cycle rather than just two arbitrary positions in it?
>
> Tom
>
>
>