[Laser] Cloud Bounce, Comm or Distance

Fri Sep 17 19:53:26 EDT 2010

Speaking for our group of Australian experimenters, I can definitely 
state that all of these concepts have been considered, and most of these 
ideas have actually been field tested over the last four decades of our 
activity. The green and blue PhlatLight  LEDs actually have a higher 
radiometric efficiency than the red, and the green LED has a much higher 
photometric (eye response) efficiency, but these have far less 
compatibility with the spectral photoresponse of silicon, which favours 
longer visual wavelengths and the near-IR. The green LED's phosphor has 
a much slower rise time than the red PhlatLight, and it has a much 
broader half-power spectral bandwidth.

When one goes to digital, even for the transmission of speech, the 
modulated tx and rx bandwidth must increase markedly. With Si P-I-N 
photodiode receivers, this inevitably means a rapid trade-off of signal 
to noise ratio against bandwidth, a redesign of the receiver's pre-amp 
to suit that broader bandwidth, and the acceptance of far smaller tx 
ranges. Avalanche photodiodes could provide part of the answer to the 
speed problem, but they are around thirty (30) times more expensive than 
the equivalent P-I-N photodiode, and they suffer from numerous noise 
sources, some thermal, and some due to their avalanche signal 
multiplication process, which is not noise free. The high 
thermally-derived dark current of avalanche photodiodes makes it 
difficult for them to equal photomultiplier sensitivity, even if 
sufficiently high current gain at low levels is achieved to overcome 
their thermal noise. The avalanche photodiodes have the advantage in 
quantum efficiency, but this may not be sufficient in itself for them to 
be a superior photodetector, overall, to a traditional photomultiplier tube.

An alternative is to use the blue-violet PhlatLight LED, in combination 
with a receiver employing a surplus photomultiplier - a superb and 
underestimated device, uniquely combining large sensitive area with 
extremely high speed, and mostly suiting the detection of the shorter 
visual wavelengths. However, in a direct line-of-sight link, the much 
greater scattering, poorer atmospheric transmission and much higher 
refractive scintillation of blue light renders this alternative 
wavelength range unattractive, at least for atmospheric optical DX.

The exception to this objection occurs where scatter, cloud bounce or 
non-line-of-sight ("NLOS") linking is contemplated. In this instance, 
the much greater Rayleigh scattering potential of shorter wavelengths 
(proportional to the inverse of the wavelength to the fourth power) 
could actually become advantageous, especially for molecular scattering 
in a cloudless sky. With most scatter links of this type, a fast and 
relatively large area photodetector (ideally, again, a photomultiplier) 
is necessary to intercept the large image dimensions of an extended 
scattering field in the receiver optic's focal plane. In this NLOS 
instance, the ideal optics would be Fresnel lenses or parabolic 
searchlight reflectors of very large aperture, perhaps in excess of a 
metre in diameter, to maximise tx optical gain and to maximise the 
scattered photon collection area for rx. The area of the sky illuminated 
should be minimised via the production of the narrowest possible tx 
beam, so that that the scatter field can be imaged on a photodetection 
area of minimum dimensions, for two reasons:

(1) A concentration of high flux in the smallest possible part of the 
sky will maximise the modulated scatter illumination against ambient 
(background) illumination.

(2) Photodetector sensitive area is always directly proportional to the 
photodetector's thermal noise, so it is always desirable that the 
photodetector dimensions should be matched, as closely as possible, to 
the image dimensions of the scatter field. The image size of the sky 
scatter field can be reduced in the receiver's focal plane by reducing 
the receiver optic's focal length as far as possible, and/or by the 
spreading of the scatter field in altitude, but not in azimuth. The 
azimuth of a given scatter field (if not for cloud bounce) will always 
maximise when two NLOS scatter stations are pointed precisely in each 
other's direction, with a minimal path distance between the two. 
However, if the tx beam spreads in the vertical direction only, and the 
azimuth of the beam remains the same, the shaft of light seen at the 
distant station will not occupy a larger sky area. It may be desirable 
to have independently controllable vertical and lateral beam divergence, 
and to mask the receiver's sensitive area to admit only the part of the 
sky including an image of the transmit beam's scatter field, which will 
usually be seen as a shaft extending vertically from the horizon in the 
direction of the transmitter. In other words, the mask will admit light 
from a vertical slot, matching the image of the scattered vertical light 
shaft.

By following these fundamental optical and practical principles, we 
contend that it should be possible to dispense with wspr, WSJT and other 
sub-noise detection systems on such a NLOS link. The recovered 
sig./noise should then be higher by many tens of dB, and speech linking 
may be quite possible by using large high-gain optical systems, accurate 
tracking, and high intensity sources such as blue PhlatLight LEDs. In 
recent times, another group of hams used WSJT to bridge about 300 km 
NLOS, by intentionally spreading a transmit beam to some 5 degrees of 
divergence from the tx, and by intercepting only a tiny part of the 
scattering field. Sixty (60) Luxeon LEDs behind 60 page magnifiers were 
used, with commercial Lumileds  catadioptic "torch" reflectors as 
secondary reflectors on each Luxeon. We would contend that these optics 
are excessively cumbersome, needlessly expensive, and very poorly 
conceived. Intentional spreading of a tx beam for scatter propagation in 
the presence of ambient is a fundamentally flawed concept. Added to that 
problem, the tiny avalanche photodiodes used for reception had 
inappropriately small collection areas for the NLOS scatter system. If 
the emitters, detectors and optical systems had been better designed and 
matched, no recourse to WSJT would have been necessary.

The usage of WSJT or similar sub-noise "digital" systems, capable only 
of transmitting call signs over periods of 10 minutes or more, have too 
often been used in circumvention of sound optical design. By the usage 
of WSJT, many optical system failures tend to have been dressed up as 
successes. I don't think that anyone would argue that progress lies in 
that direction.

In other words, there is AMPLE scope for improvement, here!

Best wishes,

Chris Long VK3AML.

David Learmonth VK3QM.

============================================

> Date: Thu, 16 Sep 2010 20:48:36 -0400
> From: n5gui at cox.net
> To: mike1 at mgte.com; laser at mailman.qth.net
> Subject: Re: [Laser] Cloud Bounce, Comm or Distance
> 
> 
> ---- Mike <mikecouture at bellsouth.net> wrote: 
> > 
> > Do we have any interest in cloud bounce for either communications or
> > distance measuring using low to medium power lasers or LED's?
> > 
> 
> 
> Your question got me to thinking ( which is usually dangerous ) about a couple of things.
> 
> The first is perhaps a subset of cloud bounce.  The classic idea of cloud bounce, to me at least, is that you see a cloud in the sky, you point your light beam at it, and then someone sees your "spot" and decides to reply.  I have frequently seen an advertising searchlight hit clouds so this does not take a lot of imagination.
> 
> There are atmospheric conditions other than puffy cotton ball clouds that might be used.  One such condition causes rainbows or fogbows.  I am thinking more of thin ice clouds.  Often in winter there are thin ice crystal clouds, usually observed with sun dogs or halos, or their equivalents from bright moonlight.  These ice crystal clouds are hard to see directly.  The ice crystals form different shaped prisms or flat hexagonal plates which reflect and / or refract light at predictable, but narrow angles.  ( That was a hint for the curious to investigate the atmospheric physics involved. )  A beam of light, whether from the sun, moon, or optical communication equipment, comes away in a cone defined by that angle.  
> 
> In order to use the reflections/refractions for communications you would need considerably more skill, or perhaps extraordinary luck, than the more typical cloud bounce.  I would not suggest anyone start on a project to use ice crystal reflections instead of, or before the puffy cloud bounce.  But maybe it is the next step ( or the next challenge after ).
> 
> The second thing was listening practice for cloud bounce, for either the puffy cloud version or the ice crystal reflection/refraction version.  In the densely populated areas, streetlights and other manmade lights pulsing at 60 Hertz and various harmonics, should be detectable on every cloud overhead.  I live in Kansas, so there are places with some pretty dark skys, so I should be able to find places where only limited sites can illuminate the clouds, puffy or icy.  Predicting and then confirming the cloud bounce, seems to be a useful practice before trying to communicate.  I am not currently working on such a project, but I thought I would pass the idea along in case someone else could use it.
> 
> 
> James
>  n5gui
> 
> ______________________________________________________________
> Laser mailing list
> Home: http://mailman.qth.net/mailman/listinfo/laser
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:Laser at mailman.qth.net
> 
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html