[Laser] Amp/phase distortion

[Andrew T. Flowers, K0SM]

G0MRF@aol.com wrote:

>
>QRSS is a great method of recovering very weak signals and it's been standard 
>practice on the LF bands for the past few years. However, it's not so ideal 
>for laser communications due to the blurring that occurs on transitions 
>between on and off periods. Don't get me wrong here, I think it's a very good 
>idea for extending the distances possible using lasers, but its limitations 
>will be far more apparent because of the scintillation of the beam in  
>disturbed air.
>

Good points here, David.  This is the same sort of problem that people 
run into when they try to use PSK31 on 2 meters (most of my experience 
with DSP and radio is with FSK441 and JT44 on 144 MHz).  I'm reminded 
that stars twinkle as their light passes through the atmosphere :-)

Would changing the modulation frequency help?  My receiver seems quite 
happy to hear in the 20-50 Hz range.  I'm thinking that at a lower 
frequency there well be less spreading across FFT bins.  That reminds 
me--I've noticed that there is more noise in the 20-50 Hz range in my 
receiver, even under completely dark conditions.  Is the PGP front-end 
more sensitive here or is this just where the diode gernerates most of 
it's noise?  I think anything below 20Hz is being filtered by by my 
soundcard. Maybe using IR would help since it's less effected by 
scattering??

Now, if you are talking about the blurring between elements in time that 
appears to "run everything together", much of that is a result of 
overlapping large FFT windows that span the gap between CW elements.  I 
think I remeber seeing something on a longwave site that talked about 
methods of "cleaning" up the screen by not overlapping windows as much, 
etc.  I haven't played with spectran enough to know if some of these 
settings can be changed or not.  Of course, the math is eventually going 
to limit what you can do.  DFCW should help this blurring a bit as long 
as one keeps the tones far enough apart to keep them from bluring too 
much.   It would also speed up trasmition time. What do you think?  Of 
course, if this works, one might as well jump to a multi-tone code like 
PUA-43 or something in which each number and letter has it's own unique 
frequency (Looks like K3PGP was playing with that a while back).  It's 
not like we need to conserve spectrum usage like we do in a 1 KHz wide 
LF band....besides, a complete contact with a 10s dot rate will take a 
very long time!!  (I might have patience for 3s dot in a contest, but I 
don't know....)

(For those on the list who have never heard of such a thing, DFCW stands 
for "Dual-Frequency CW".  Instead of on-off-keying of a single tone, two 
tones are used: high="dah" and low="dit".  The result is that you can 
eliminate silence within morse code letters and use shorter gaps between 
words.  It's kind like "morse code meets teletype" if you want to think 
of it that way.)

I'll try to get two stations assembled in the near future and actually 
try some of this for real.  Right now I just need to get the 
transmitters operational.  I'm going to get 555 timer and play with it 
this evening.  We'll see how well it does.  Unfortunately, I live in 
downtown Rochester, NY, and I can't point at the clouds without pointing 
into a sodium lamp, and creating atmospheric distortion in my apartment 
is kinda tough....I'm really glad I can afford to build this stuff--I 
have less than $20 invested so far!  Beats getting on 47 GHz on my grad 
stipend :-)

I'm curious what other people have discovered?  I've received e-mail 
from a few folks who have dabbled around with this stuff...

Andy K0SM/2