[NLRS] Light wave communications

[Doug Reed]

Hi gang.

I've sent a number of emails in the last year with information about
light wave (optical) communications. In the past we'd have been
talking about using lasers for communications, and the ARRL contest
rules used to be written that way. The current rules specifically
mention LEDs and still require electronic amplification on the receive
side.

"1.12. Above 300 GHz, contacts are permitted for contest credit only
between licensed amateurs using mono-chromatic signal sources (for
example, laser and LED) and employing at least one stage of electronic
detection on receive.  Laser usage is restricted to ANSI Z136 Class I,
II, IIa, and IIIa (i.e.; output power is less than 5 mW)."

Since most local and regional governmental organizations prohibit
shooting lasers into the air, LED light sources are a much "safer"
bet. In addition, they are easy to expand and collimate for best
reception at a distance. The 2005 amateur radio optical communication
record in Australia was based on 1 watt red LED diodes and page-size
Fresnel lenses and reached over 100 miles.
<http://www.modulatedlight.org/Modulated_Light_DX/MODULATED_LIGHT_DX.html>

The large beam diameter also reduces the amount of signal loss due to
scintillation in the atmosphere. The web pages by Clint KA7OEI are the
best US source of information on amateur optical comms that I've
found....
<http://www.modulatedlight.org/optical_comms/optical_index.html>

Many more links on this web page: <http://www.aladal.net/toast/comlinks.html>

Within the last few months, some German hams tested a fast-scan ATV
video link over a 50 mile path using 20mw IR lasers and photo
detectors. They couldn't use visible red light because they were
shooting along a frequently used air corridor between the two
sites....
<http://www.darc.de/index.php?id=24358>

And while line-of-sight is a good idea and was required for the
longest distance records, it isn't a requirement! Just like we do rain
bounce and snow bounce at 10GHz and 24GHz, for light comms, we can use
cloud bounce and atmospheric scattering to extend range beyond your
local obstructions. If you look around a bit you will find
documentation of laser light bounce and scatter tests that were run
over 15 years ago.
<http://www.k3pgp.org/laserscatter.htm>

I would tend to say that cloud bounce is going to be short range if
you think about low-altitude storm clouds. But if you think of those
high wispy clouds you get on a clear summer day, those should be good
for some pretty long distance....

"Scientists at the International Telephone and Telegraph Federation
laboratories in New Jersey have tried using clouds as the reflecting
medium. They found that the typical cumulus clouds can scatter the
beam to a receiver more than 150 km. from the transmitter."
    From "LASERS, Tools of Modern Technology", 1968, Page 82.

A better choice might be atmospheric forward scatter from dust and
other particulate matter in the air when you are shooting near the
horizon. If nothing else, there are ice particles in the troposphere
30 miles up.... The K3PGP link mentions tests he ran out to 26 miles
using digital modulation of a 5mw laser light beam in 1997.

Regarding the choice of LED diodes, if you look on Ebay or in any
catalog, you will find that high-power LEDs are available in just
about any color you could want. Most LED comm work has been done using
red LEDs near the same 650nm wavelength generated by most laser light
pens. I'm sure you've all read stories about people getting arrested
for shining a laser pointer at car drivers or aircraft pilots. How
would you like to be the one shining a very bright 8" diameter beam of
red light into the air from our favorite Mounds Park hill above the St
Paul downtown airport? How long do you think it would be before the
police arrived? Other than for some short range testing, I think I'd
like to avoid visible light if I can.

If you avoid visible light, your next choices are infrared: 850nm,
940nm, and 1550nm. 940nm is the middle of an absorbtion peak so we
don't want to use it. 850nm is the most commonly used wavelength for
amateur light comms. 1550nm is the wavelength most often used for
"free space optical data links" and might be the best wavelength
overall, if you can find the parts. Most 1550nm equipment on Ebay is
for fiber optic use. But 1 watt and 3 watt 850nm LEDs are easy to find
on Ebay at quite reasonable cost. The primary difficulty with
non-visible light is trying to verify you've adjusted the focus
correctly....

Another good thing about using 850nm infrared light is that the
receiver photo detectors can be purchased with a built-in infrared
filter to block at lot of the ambient visible light interference. If
you check the catalogs you will see that most detectors are available
with and without the "F" filter option and the only difference between
them is the F model is restricted to infrared light and the visible
spectrum is attenuated. That improves your receiver sensitivity, just
like putting a bandpass filter ahead of your HF/VHF radio allows it to
shrug off strong out-of-band signals.

What I'd like to suggest is that if you are interested in playing
around with optical communications, you plan to use 850nm in the long
run, although I admit it is a LOT easier to start with common red LEDs
when first playing around... If you start with a red LED for transmit
and a visible light photo detector, you can change to a 850nm IR LED
very quickly and either put a IR filter ahead of the photo detector or
replace the detector with the "F" version. (Did you know that the
black portion of a 35mm black & white film negative makes an excellent
IR filter?) But don't forget that IR light will focus at a different
distance than the red light so you will have to adjust the diode mount
when you switch to the IR diode.....

As for the communications mode, perhaps the easiest is CW with on-off
tone keying of the TX LED. That is effectively what Gary WB0LJC and
Bob W0AUS and everyone else were doing 20-40 years ago with
chopper-wheel modulated Helium-Neon lasers..... Most current
experimenters are using some sort of voice modulation: baseband, PWM,
sub-carrier FM, or SSB. What I prefer is to stay with a FM mode that
can use on-off keying of the LED for maximum signal amplitude.

If you are already playing with any of the sound-card software
programs like FLDIGI, you already have most of a very capable digital
communications platform running on your computer. You could very
easily connect it to a LED transmitter and photo-detector receiver to
do optical communications over a very respectable distance.... Other
experimenters have shown that even a simple optical system using a 4"
lens can make Q's longer than 30 miles.
<http://www.earf.co.uk/nanotrx.htm>

I'd like to consider using PSK31 or a similar digital mode for the
modulation. We have the whole response band of the photo detector to
work with, but lets limit ourselves to the range below 25KHz. This
gives us 20+KHz of bandwidth and would let us have a lot of QSO's
going on. With software like Multi-PSK or PropNETPSK, your detector
would pick up any beacon signals within its field-of-view, and the
software would decode and display them. PropNETPSK even has a robot
QSO mode where it beacons CQ CQ CQ DE N0NAS N0NAS and will
automatically work any station that replies. Or if robot mode is off,
it still transmits beacons on a regular basis and logs any beacons it
hears.

For a real silly-season idea, how about mounting your light comms
equipment to an AZ-EL rotor, run the ADSB software on a DVB-T dongle
to get real-time location info on airliners up to 250 miles away. Plug
the location into the 10GHz rain scatter program to predict when the
plane is in position for forward scatter to a distant station. Let the
software point the rotor and start sending a beacon.... This might
even work pretty well with sound-card meteor scatter software.... We
don't have to hit the plane directly, all we need is to hit the
contrail behind the plane for a few minutes in order to make the Q.

By the way, if you want to try this for 10GHz airplane bounce,
coupling the ADSB plane beacon and the rain scatter program would
probably work to predict when a plane would be in the desired path.
You can find a good site that talks about "Aircraft Enhanced
Propagation" at this site:
<http://www.qsl.net/vk3bjm/ads-b.htm>

If you search some of the ADSB software links, you will find that the
ADSB software can be configured to forward its "catches" on the
Internet. There are web sites that collect the info like the APRS
sites and it is possible to get data from remote user's so this could
be a full-blown path prediction system.... I know there was discussion
in the past about trying 10GHz plane scatter but the problem was
knowing when the plane would be in position. With the ADSB receiver
software, that problem goes away..... Just something to think about...

I was using ADSBscope for display software, with the ADSB# (ADSBsharp)
receiver software for the R820T DVB-T dongle. Less than $20 worth of
hardware.... It will work quite a bit better with an outdoor
antenna.....
<http://www.sprut.de/electronic/pic/projekte/adsb/adsb_en.html>

73, Doug Reed, N0NAS.