MODULATED LIGHT DX:
Long-distance atmospheric optical comms. pages by Chris Long and Mike Groth VK7MJ:
**Double click on these BLUE titles to reach further references:
Mike Groth's overview
of optical comms. theory: "PHOTOPHONES REVISITED".
OPTICAL RECEIVERS -
ELECTRONIC DESIGN.
Chris
Long's
"OPTICAL COMMUNICATIONS FOR THE AMATEUR", on optics etc.
"THE LUXEON":
new light of
hope for optical communications.
On
light modulation
by ULTRASOUND in a liquid cell.
Chris Long's account
of his OPTICAL COMMS. EXPERIMENTS, 1968 - 80.
BIBLIOGRAPHY OF
OPTICAL (AUDIO) COMMUMICATIONS, 1878 -
1940.
Optical comms. before
electronics: 'THE HELIOGRAPH'
(written c1899).
Optical
through-the-air Commications page by KA7OEI.
Project Red
Line - the 1963 long-distance laser experiment - and
photo gallery.
FOR THE 167 km COMMS. RECORD IN TASMANIA - REFER BELOW:
104 miles by 'LUXEON' -
An alternative to lasers for optical comms. - Tasmania, 19 February 2005.
by Mike Groth VK7MJ and Chris Long.
On the evening of 19 February 2005 the authors spanned a 167 km (104 mile) path between Mount Barrow's South peak and Mount Wellington in Tasmania with audio modulated light beams. This was the culmination of about 35 years of intermittent experiments by the two authors of this web page, initially individually and later in association. The suggestion to try the Barrow-Wellington path originated in a chance comment to Chris Long by Jason Reilly VK7ZJA, that Mount Wellington was visible from the top of Barrow. Jason is a technician for Optus, regularly visiting the transmitter site on Mount Barrow, otherwise we would never have known that crucial piece of information! Both summits are above the 1200 metre contour, and the earth's curvature on the land between only accounts for approximately a 600 metre bulge, with reasonable clearance all along the path. We were surprised to find that the expected rapid optical fading cycles caused by atmospheric turbulence over this long path were not too severe, presumably because the effect rapidly diminishes with elevation.
By extreme luck, our initial Saturday night DX test from South Barrow occurred following a day of rain. This extinguished all open-air fires and washed most of the dust out of the 104 miles of atmosphere between the optical stations. The visual range on the night selected was outstanding. News of our subsequent success with the optical link rapidly spread to the Internet. Reports, photographs of the equipment, audio grabs etc can be found on the pages of the Radio and Electronics Association of Southern Tasmania, which hosted a small celebration in Hobart on 23 February 2005 (photographs of the celebration are included):
http://www.reast.asn.au/optical.php
Further photographs of the Northern end of the link (South Peak of Mount Barrow) on the 19th February can be found at the web site of Jason Reilly VK7ZJA who assisted in the Launceston arrangements and shot the photos. These include pictures of Joe Gelston VK7JG (refer picture DSCF0183.jpg) who was kind enough to transport us to the summit in his four wheel drive van on two occasions, dodging numerous nocturnal marsupials, unlocking the necessary gates and scouting out sites that contributed significantly to our success. At Mount Barrow, we had two complete optical transceivers fitted with PIN photodiodes and Luxeon LED light sources. Both optical transceivers used fresnel lens collimators 19 cm by 25 cm. The two transceivers were spaced about two metres apart on Mount Barrow, which we successfully used to provide a degree of spatial diversity reception.
|
Click here for
optical |
|
Our chosen optical transmitting site (see above) was right at the base of the Australian Broadcasting Corporation's TV transmitter tower on the South Barrow peak. Refer to Jason Reilly's web page for further photographs of the optical equipment and the panoramic views from the South Barrow peak prior to sunset at:
http://members.optushome.com.au/~jason.reilly
The authors would like to thank the Tasmanians who provided local lodgings for Chris Long, who brought his gear from Melbourne for twelve days specifically for this DX attempt (16 Feb to 28 Feb 2005). Their hospitality greatly reduced the overall expense of the exercise. In Launceston (Kings Meadows), Chris stayed with John and Marje Corrick; while in Hobart, Chris stayed with Alison Melrose (Bellerive) and Peter and Libby Mercer (Cascades). In any amateur endeavour, the participants are entirely self-funded and costs therefore become a major concern. These Tasmanians have maintained their island's deserved reputation for hospitality!
For the information of people living outside Australia, the island of Tasmania is Australia's Southernmost state, located between latitudes 40 and 44. It's about 300 km South of the mainland state of Victoria, across the notoriously treacherous waters of Bass Strait. Tasmania enjoys a temperate and cool maritime climate similar to Southern England or France, with a greater and more reliable rainfall than most of Australia, the result of a mountainous topography intercepting the circumpolar 'Roaring 40s' winds.
The path of our optical communication beam (map on the lower right) crossed farmland with minimal industrial air pollution, spanning most of the length of the island.
|
Details of Tasmanian contact, 19 Feb. 2005: At the Mount Barrow end Joe VK7JG, Phil VK7JJ, Jason VK7ZJA, David VK6YA/7 and Chris Long were present; while at the Mount Wellington end Mike VK7MJ and Justin VK7TW manned the mountaintop station. Communication was simple audio amplitude modulation of 1 watt Luxeon LEDs, in full duplex, and with about 10 to 12 dB s/n (max). This is certainly an Australian distance record for optical communication, and it is a world record for two-way audio-modulated optical communication using non-coherent light sources. It indicates that lasers may not necessarily be the most appropriate technology for atmospheric optical DX. By comparison, the currently accepted North American amateur record for two-way amateur laser audio communication using the same frequency as ourselves, red light at 474 THz or 630 nm approx, is only 92 km (57 miles) between WA6EJO and K6MEP on 9 June 1991 - approximately half the distance achieved in Tasmania. On 21 September 1997, a contact in Arizona on HeNe lasers involving KC7AED and N7VUB at Smith Peak; and WB7VVD and KC7PCV at Four Peaks spanned 192.4 km. However, this was MCW only, not voice. |
|
|
|
------------------------ A record one-way audio DX contact was achieved by Jack Pattison W6POP, Duane Erway W6KAQ, Robert Legg W6QYY and others way back on 3rd & 4th May 1963. It was a 190 km (118 miles) temporary audio link set up by the EOS amateur radio club of Electro-Optics Systems of California. The club included 16 licensed hams and 6 interested non-hams. They used a 125 µw HeNe laser (474 THz or 632.8 nm red light) entirely fabricated by themselves, set up in the San Gabriel Mountains (alt: 7000 feet) near Wrightwood in Southern California. The laser was energised by the output of a 28.62 MHz a.m. (voice) transmitter. Their receiver was a 12.5 inch reflecting telescope and photomultiplier with an S-20 photocathode, providing only 4% quantum efficiency at 632.8 nm. This was erected inside a tent, 5000 feet up in the Panamint Mountains, West side of Death Valley, near the town of Ballarat, California. The beam occupied a South-to-North path across the Mojave Desert. In 1963, the ARRL considered that this was
'not a transmission using the amateur radio spectrum', therefore
not an 'amateur radio' contact. Attitudes have changed
since...
This benchmark DX record was unchallenged for 25 years. For more details about this historic
communication, see the "Operation
Red
Line" web page. ------------------------ |
|
|
|
Another distance record for atmospheric optical DX, set by KY7B/7, WA7CJO and WA7LYI on 8 June 1991, used violet light generated by an unspread 15 mW HeCd laser on 678 THz (442 nm). The photomultipliers used for receiving are devices of unrivalled sensitivity, but work best at the short wavelength (green-blue-violet) end of the visible spectrum. Violet laser light had the PM's working efficiently, with fresnel lenses 46 cm square for reception. However, in refracting more than red light, violet is subject to extreme atmospheric scintillation and absorption. This all-time atmospheric optical DX record spanned 248 km (153.97 miles), but was set with MCW tone only, not with voice modulation. This distance probably could not be achieved in Australia, as we have no mountains high enough to provide a lengthier line-of-sight. |
|
|
|
------------------------ More recently, the the bounds of optical DX
were
pushed further by KA7OEI and K7RJ in Utah. On 18 August, 2007, a
two-way voice contact was completed
over a distance of 172.27km (107.04 mi) under very poor seeing
conditions. On 3 September, 2007, taking advantage of clear air,
this
path was re-done (this time, a distance of 173.34km or 107.09 miles)
and excellent
results were obtained using transcivers that utilized Luxeon III LEDs
and PIN photodiode receivers: On this occasion, a laser beam
collimated with a 20.3cm diameter (8 inch) reflector telescope as well
as a cheap, off-the-shelf laser pen were used to span the same
distance, but with significantly greater scintillation. Read
about these two contacts on the "107
mile Optical QSO" page.
On 3 October 2007, the bounds were pushed out yet again by the Utah team with a 2-way morse contact between Swasey Peak in central Utah and a site near George Peak in northwestern Utah - a distance of 278.6 km (173.1 miles) and a one-way transmission was completed using voice: Had the air been slightly clearer, 2-way voice would have been possible. Read about this contact on the "173 mile Optical QSO page. |
|
------------------------
After writing the above, Jim Whitfield, N5GUI, informed us that the current all-time record for optical communications using Morse code via light was a greater distance than any of the records listed above. It was non-electronic, indeed it was not even electric. It was by heliograph - a mechanically tilted mirror arrangement reflecting sunlight to provide Morse 'keying', and the 'signal' was received by the human eye. This record was established by the United States Army Signal Sergeants, from Uncompaghre Peak, Colorado; to Mount Ellen in Utah. The distance? 183 miles! The date? 1896 !!!
With 109 years of optical and electronic development, surely we should at least equal that record today with speech modulation - or is atmospheric pollution now so bad that this can no longer be achieved? Who will take up the challenge, 'walk up to the plate' (or the mountain) and do it?
------------------------
Almost all of the electronic optical comms DX records set during the last three decades have employed plastic fresnel lenses for receiving, and our own tests have additionally employed fresnel lenses as transmitting 'antennae'. No other type of lens can provide such enormous apertures, light weight and freedom from spherical aberration at low cost.
For an excellent monograph on the theory, manufacture and usage of fresnel lenses, refer:
http://www.fresneltech.com/pdf/FresnelLenses.pdf
Acrylic fresnel lenses 39.5 cm by 39.5 cm, 2 mm thick, with a focal length of 33 cm (part number #A395a) are available from the 3dLens company in Taiwan for US $28.60... see:
http://www.3DLens.com/Large_Fresnel_Lens.html
Try comparing this with the cost of a 40 cm diameter glass telescope achromat or reflector objective..... (!) In reducing the various optical communication components to a practical system, one aims at maximum overall sig/noise for minimum dollars spent. Theoretical perfection is only a secondary consideration. Sure, we'd all like to 'play' with telescope objectives, accurate to 1/4 wavelength, diffraction limited, with steel 'I beam' mountings set into 20 metres of concrete and with directional guidance micrometers accurate to 0.0001 microradians. But how would one get these up a mountain in a car, or mount them on a roof or even on a tree? Fortunately, while this precision equipment is all very nice, it's not really necessary. I'm sure we'd all like to have the same income as Bill Gates, also...(!)
Why not lasers?
This question raises the ire of folk who have spent countless thousands of dollars in assembling high-powered lasers, large-aperture optics with surfaces accurate to 0.25 λ, and aiming mechanisms accurate enough to fire a thread through the eye of a needle at a distance of twenty kilometres. Sources of coherent light, ie. laser sources radiating plane parallel monochromatic beams of light with their wave fronts in phase, have traditionally been regarded as the only serious contenders for atmospheric optical communications. Unfortunately, few experimenters who have the electrical knowledge to use lasers are simultaneously fully aware of their problems of transmission through the atmosphere.
Lasers have many practical shortcomings, even if it were legal or desirable to shoot these beams indiscriminately across populated areas where eye hazard is always a concern. First and foremost, most gas lasers have a limited shelf life. Monatomic helium has such a small atomic radius that this gas will gradually leak through the glass walls of any Helium-Neon laser tube, even in storage. The tiny helium molecules can actually seep through the spaces between the glass molecules. This would be acceptable if individual HeNe plasma tubes were cheap or simple to replace, or indeed if HeNe lasers could easily be internally modulated. Even the least expensive gas lasers are not cheap, and individual laser tube types, particularly from disposals sources, are frequently difficult or impossible to replace when they fail. Furthermore, the modulation of a HeNe plasma tube via its power supply can only be achieved at a base band audio rate, owing to the finite ionisation time of the gas. That type of internal modulation can only be applied with very small modulation percentages before the tube drops below its lasing threshold, becomes non-linear or exceeds its dissipation ratings.
Diode lasers are cheaper, easier to modulate internally, and have a very fast rise time. However, they are electrically fragile, the coherent percentage of their output flux is limited, and their radiation is not inherently collimated. Without external optics, diode lasers emit radiation in a roughly ellipsoidal cone, over about a 30 degree angle perpendicular to the junction, and over a 10 degree angle parallel to it. For this reason, diode lasers are almost always fitted with a small (usually 1 cm or less) collimation lens. Even then, the cross-sectional flux density of most diode laser beams is unevenly distributed. This provides problems in propagation through the atmosphere, particularly when beam spreading through a large collimating lens is attempted to reduce the dispersal resulting from diffraction.
Let us be blunt: coherent light is incompatible with the turbulent atmosphere. It will only transmit a few hundred metres horizontally before the coherent wave fronts begin to break up as they encounter constantly moving cells of higher and lower density air. Transmitted over any significant distance, the beam will suffer from deep, rapid scintillation or fading. In atmospheric transmission, coherent light beams can be far less capable of carrying recoverable modulation than an equivalent beam from a non-coherent source. Atmospheric phase and amplitude noise mostly renders heterodyne detection via a local laser oscillator impossible.
Optical frequencies possess one markedly different characteristic to communication at radio frequencies, in that the transmission medium can no longer be regarded as transparent and unchanging. While laser light would certainly be the choice if we lived in a vacuum, the atmosphere is a turbulent, dynamic medium containing moving air cells of high and low density as well as high and low temperature. The amount of dust and particulate matter in the air varies constantly, as does the amount of water vapour, clouds and fog.
The graph shown below is representative of typical results over a 28 km path near sea level. This readout is from a 1 kHz tone-modulated beam incoherent light beam from Luxeons transmitted through wide-aperture fresnels. The graph shows 0.4 seconds of the constant-amplitude tone after transmission through 28 km of air. With a laser source suffering de-cohering noise, the fading would have been much deeper and probably more rapid:
Number of views since 1/2010:The atmosphere's 'turbulence bubbles' or 'cells' move through any optical communication beam with the prevailing wind, causing instantaneous optical changes in any light beam path. Light passing through these moving air masses will encounter refractive indices that continuously vary, typically causing a fast 'flutter' fading or scintillation of the optical signal. Coherent light sources exacerbate this problem, as the loss of the coherent wave fronts through atmospheric turbulence causes a de-cohering noise. The wave fronts variously add or cancel in random ways dictated by the varying refractive index of the moving turbulence bubbles. The problem rapidly increases with distance, and over a 20 km path with a beam 3 cm in diameter the cyclic fading rate can be as high as 500 Hz. These atmospheric limitations were discovered in the initial unguided laser communication systems of the late 1960s, and they provided the major impetus for the abandonment of atmospheric links in favour of optical fibre technology in the 1970s. Since then, of course, optical fibres have become the dominant technology.
The result was that for most of the 1970s and 1980s the development of atmospheric optical links lay relatively dormant, with only the occasional short distance link receiving publicity - together with some attempts at longer-haul optical comms by amateurs like ourselves. These amateur links generally used relatively narrow modulation bandwidths, typically base-band audio amplitude modulation of the light source.
A revival of interest in short distance (less than 3 km) optical communication systems has followed the rapid upsurge in the need for broadband LAN computer linking, stimulating research into the unique digital optical communications problems encountered. The most difficult problem is the rapid cyclic optical signal fading caused by moving atmospheric turbulence cells. One of the first papers to measure and analyse the problems of relatively long distance atmospheric optical communications was:
B G King, P J Fitzgerald and H A Stein: "An Experimental Study of Atmospheric Optical Transmission" in Bell System Technical Journal Vol.62, No.3, March 1983 pps. 607 - 629.
While we can find no freely available Internet download of this paper, its associated American patent 4491982, awarded to J C Candy and B G King of AT&T in 1985, can be found on the Internet at this address, regrettably without the major analytical work that led to it:
United States Patent 4,491,982
This was, to our knowledge, one of the first laser communication systems to use very large fresnel lens arrays for receiving. It had an accurate beam directing system to overcome the diurnal variations in the vertical micro-bending of the beam, particularly at sunrise and sunset when the earth was changing in temperature with respect to the air above it.
In recent research papers, scientists and engineers have suggested various means for reducing the de-cohering phase noise and rapid amplitude fade caused by atmospheric turbulence in laser comms systems. This paper by X Zhu and J M Kahn from the 2002 transactions of the IEEE provides a detailed analysis of the problem, and it indicates the potential value of diversity reception:
http://www-ee.stanford.edu/~jmk/pubs/trans.com.ml.det.turb.pdf
Indeed, we can vouch for the practical value of optical spatial diversity reception. Our two optical receivers on Mount Barrow, spaced about two metres apart, were independently amplified into separate loudspeaker systems. Standing between these loudspeakers and listening to the transmitted beam via both speakers, in stereo, one could copy the transmitted audio from Mount Wellington far more easily than by listening to either receiver on its own. The timing of the fading nulls on each receiver was almost completely mutually exclusive, even with the relatively close spacings of the receivers. This suggests that to alleviate the problems of localised fading, two small transceivers spaced some distance apart may be more effective than a single large optical transceiver.
The atmospheric turbulence fading problem also affects higher microwave frequencies as this IPN/NASA paper by Ho and Wheelon from August 2004 indicates:
AtmosScintillationAtMicrowaveFreqs.PDF
One solution for atmospheric de-cohering noise from laser sources involves placing a thin diffuser over the laser output window. This partially de-coheres the source in a controlled, constant manner before transmission, thereby avoiding deeper wave-front cancellations and fade drop-outs.
Sounds unlikely? Refer these papers by Korotkova, Andrews and Phillips, published as recently as 2003. The maths are rather fearsome but the underlying principle is clearly stated and its effect is proven, on paper and in practice. In particular, note the graphs of scintillation index for coherent and incoherent beams in the first paper on this list - 'OptEng43', 'Model for partially coherent Gaussian beam in atmospheric turbulence [...]'. The severity of scintillation is clearly proportional to the degree of beam coherence, an undesirable characteristic of atmospheric laser communication:
http://pegasus.cc.ucf.edu/~okorotko/OptEng43.pdf
http://pegasus.cc.ucf.edu/~okorotko/OptLett29(11).pdf
http://pegasus.cc.ucf.edu/~okorotko/SPIE4821.pdf
http://pegasus.cc.ucf.edu/~okorotko/SPIE4976.pdf
http://pegasus.cc.ucf.edu/~okorotko/SPIE5160.pdf
The following papers by the formidable Olga Korotkova may be of interest if you're contemplating optical POLARITY modulation for conveying data over an atmospheric path. They deal with the effects of the atmosphere on the angular polarity of a polarised light beam transmitted through atmospheric turbulence:
http://pegasus.cc.ucf.edu/~okorotko/OSA_invited.pdf
http://pegasus.cc.ucf.edu/~okorotko/WRM14.pdf (Mohamed Salem, Olga Korotkova, Aristide Dogariu and Emil Wolf)
Another approach involves accepting a certain degree of signal loss through fading, and using digital modes of transmission with Reed-Solomon, Hamming, Golan or other error correction codes. These either permit the data stream to be interleaved with plenty of information redundancy, increasing latency while the fading drop-outs simply reduce the channel speed. This mostly results in an effective (though not an actual) reduction in bandwidth. Alternatively, the information can be sent in bursts shorter than the average predicted fade cycle. The fade cycle is faster and deeper for beams of small cross-sectional area, which strongly augurs towards the usage of large-aperture optics. Papers by Zhu and Kahn dealing with the interleaved or data burst approach to optical fading may be found at:
http://www-ee.stanford.edu/~jmk/pubs/trans.com.smc.turb.pdf
http://www-ee.stanford.edu/~jmk/pubs/trans.com.coding.turb.pdf
http://www-ee.stanford.edu/~jmk/pubs/spie.01.pilot.turb.pdf
See also:
X Zhu and J M Kahn: "Mitigation of Turbulence-Induced Scintillation Noise In Free Space Optical Links Using Temporal-Domain Detection Techniques" in IEEE Photonics Technology Letters, Vol 15, No 4, April 2003 pps 623 - 625. Download the paper at:
http://www-ee.stanford.edu/~jmk/pubs/turbulence.ptl.11.02.pdf
Also refer:
S M Haas, J H Shapiro and V Tarokh: "Space-time codes for wireless optical communications", in EURASIP Journal on Applied Signal Processing, 2002-3, pps. 211 - 220, copyright © Hindawi Publishing Corporation 2002:
DigitalCodingAvoidingOpticalFade.pdf
and
http://rleweb.mit.edu/pr2002/files/pdfs/20.pdf (RLE Progress report 145)
In the next section of this web page (currently under construction) we will examine the latest type of non-coherent source to become available for optical communications - the new 'Luxeon' ultra-high output LEDs which facilitated our recent record-breaking DX tests.
These pages were originally posted on the "Modulated Light DX" page at Tony Sanderson's Bluehaze website and have been updated and made available here with the cooperation of the authors.
Page content last revised: Sat 25 Feb 2006 with minor revisions Jan 2010.
Go to the "Modulatedlight.org" main page
