FreeSpaceOptics
From Scratchpad
back to http://scratchpad.wikia.com/wiki/Sasecurity
[edit] Free space optics
Url of this page: http://bit.ly/oQOOO. Opensource Free space optics(FSO) from Ronja - http://ronja.twibright.com/mesh.php is the most cost effective device for the creation of a hotspot between multiple houses instead of MeshNetworking. The Ronja has range of 1.4km and throughput of 10megs with 13cm Jewelers Loupes. With a loupes of 11cm on Tx and a 20cm Fresnel lens for Rx the distance is 2.5km as per TransimpedenceAmp and RonjaFpga. See http://www.modulatedlight.org/optical_comms/optical_index.html for 24km link using larger Fresnel lenses. The latency is as the same as with fiber switches, nearly unlimited repeater nodes can be created. This specifically can't be done with MeshNetworking. Lets presume there are 400 houses in a CommunityBlockNetwork. Create as many FSO hotspots as possible. The more hotspots the less DsLam and MeshNetworking nodes will have to be installed reducing the cost of the network considerably. Link these hotspots with a TelephoneNetworkRollout or WiMax on 2.4 and 5.8ghz. ElectrConsultant can be used to extend the design and improve it with FpGa logic for a max of 100megs with a LED.
Designate one home as the hotspot and lets presume ten surrounding houses have a short range 1km LOS to this hotspot. Install ten FSO links at multiple points on the roof. Connect these FSO links which each have an Ethernet port to a switch router. Inter-building nodes are created in the same way inside a CBD area connected to CctvCameras for surveillance. Note the scalability of a line-of-sight transmitter. To set up a mesh, one can install multiple tubes at one site - http://ronja.twibright.com/mesh.php - while avoiding the interference that would occur with most WiFi antennae. Create a network by cabling 50 homes in a street with DsLam. Install 10Ronja's on each house for a total of 500 nodes that with just three repeater Ronja's will have a radius of 10km. If these 50-homes hot-spots in turn are connected with 1Gig FSO links we could cover the entire Gauteng area in a matter of months providing everybody with UnlimitedBandwidth. DslamRonja combination. Or connect all ten the Ronja's on the roof via a POF(plastic over fiber) system from http://www.firecomms.com, see FiberSwitches. POF (Plastic optical fiber) replaces fiber for 155Mbit/s links and is an alternative to using a DsLam due to the high cost of copper. Using the TZA3033, TZA3034 from TransimpedenceAmp any distance of http://en.wikipedia.org/wiki/Plastic_optical_fiber is created with RC-LEDs(resonant cavity leds) using a normal copper based Ethernet switch. Repeater TransimpedenceAmp nodes are created every 50m, the limit of POF. In commercial laser diode based systems EDFA (Erbium doped fiber amplifiers) are used to receive the incoming optical signal.
[edit] Short range 10Gig links with laser
Beyond 50m scintillation effects requires OpticPatents methods. Below 50m using the Terabeam http://www.patentstorm.us/patents/6868236/fulltext.html idea scintillation doesn't have any effect on laser transmission and no linear actuators or mirror deforming techniques are needed. These short range links is an alternative to cutting the roads in the Norht-South direction as per TelephoneNetworkRollout with MicroDuct under the roads. Each home on such a 10Gig backbone functions as a repeater node receiving data at the front and exiting at the back of the home. The two FSO nodes on either side of the home are bridged with fiber inside the roof. Kilometers of houses can be connected in a line, FSO is like a fiber switch, there are no latency issues. From this backbone 2.5km, 5km etc. LED bases Ronja systems and MeshNetworking are integrated. Hundreds of houses peer of this backbone.
[edit] Extending Ronja to 10km
Url of this page: http://bit.ly/oQOOO. Patent: [url]http://www.patentstorm.us/patents/6868236/fulltext.html[/url] Terabeam : In this patent laser diode light is collimated by a collimating lense and received by a receiving lens. EDFA (Erbium doped fiber amplifier) are used to receive the laser diode light. Seven fibers are spliced into one dual-mode fiber(200micron).
Method for combining multiple optical beams in a free-space optical communication system. Multiple optical beams of the same wavelength reduces atmospheric effects (Siedrovich SPIE Jan.2002) and increases the optical power for extended distance transmission. The patent idea can be used to increase the Ronja FSO system [url]http://ronja.twibright.com/mesh.php[/url] from 1.4km, 2.5km to 10km. Combine multiple TransimpedenceAmp [url]http://luxeonstar.blogspot.com[/url] into fiber pigtails from http://thorlabs.com RonjaChips(http://bit.ly/2hKngp) focus the combined 10meg LED based design through a loupes (13cm)Tx and receive the optical signal with a Fresnel lens of any size(20,30,40cm). A single Lumiled HPWT-BD00 as per [url]http://luxeonstar.blogspot.com[/url] achieves 2.5km with Tx 11cm Loupes and 20cm Fresnel lense Rx. With 13cm Loupes at both Tx and Rx the distance is 1.4km. The larger the lenses the longer the optical distance. There aren't any jewelers loupes available larger than 13cm for Tx(transmission), but Fresnel lenses(Rx receiving) can be obtained in any size from OpticalManufacturers. By combining multiple FiberLed resonant cavity leds or Lumileds HPWT-BD00 as presently used in RonjaChips(http://bit.ly/2hKngp) the optical power is increased by orders of magnitude for perhaps a 10km link using LEDS alone which don't have the atmospheric effects that inhibit laser based systems and they are eye safe. http://www.modulatedlight.org/optical_comms/optical_index.html "....Unfortunately, such a "point-source" LED does not exist and the apparent size of the LED's emitter is the major limitation for the reduction of beam divergence...." which was solved with the Terabeam patent.
[edit] Focusing multiple LEDS via a Fresnel lens unto a fiber cable for a 400Mbit link over 10km
Multiple 400Mbit/s resonant cavity LEDS are stacked onto a heatsink, transmitting via a collimator,FresnelLens to the Phillips TZA3023(transimpedance) with TZA3044(post amplifier to PECL logic). The TZA3023 allows 1.25Gig througput see TransimpedenceAmp.
http://www.patentstorm.us/patents/4902089/fulltext.html which expires 2010 relates to a solar ray collecting device in which the sunlight, focused by the use of a FresnelLens, is effectively guided into a fiber optic cable. Adapt this patent and use multiple RonjaChips http://www.alldatasheet.com/datasheet-pdf/pdf/228353/LUMILEDS/HPWT-BD00.html stacked side by side onto a heatsink, then focus this LED energy through the FresnelLens fiber combination to create a point LED light source. This fiber light source in turn is focused([url]http://en.wikipedia.org/wiki/Collimated_light[/url]) through a collimator(http://en.wikipedia.org/wiki/Collimator) Fresnel lens combination to transmit an optical signal over 10km as per http://www.modulatedlight.org/Modulated_Light_DX/OpticalComms4Amateur79.html. Note that modulatedlight.org said that a point light LED source isn't possible, the patents mentioned solved the problem.
Another idea is to focus the FresnelLens energy unto a CPC (compact parabolic concentrator) and then unto fiber optic cable as per http://www.patentstorm.us/patents/6384320/fulltext.html. Mutliple fiber cables combined are focused on to the http://en.wikipedia.org/wiki/Collimator.
Or Stack multiple LEDS onto a heatsink and shine first into a CPC from http://www.globalspec.com/FeaturedProducts/Detail/Compound_Parabolic_Concentrator/87379/0 or http://www.edmundoptics.com/Onlinecatalog/displayproduct.cfm?productID=3093 , the CPC then focus this LED energy unto a FresnelLens which guides this into a fiber optic cable. Multiples of these cables can be spliced together to increase the energy of the fiber point source as per http://www.patentstorm.us/patents/6868236/fulltext.html.
[edit] Way to extend a USB datatream
A novel way to extend a USB camera stream with fiber cable from the Linux compatible BT878 Connexent chipset(Startech SVID2USB2) is to couple the 5volt PECL USB logic to a LED or laser driver and interface with a OM5804 Receiver demoboard for 155/622/1250 Mbps prototyping platform available for purchase from China here - http://bit.ly/BmDuS. After getting it to work a more compact board can be made(4 or 6layer). The board uses the Phillips TZA3023 chipset with its datasheets and description at TransimpedenceAmp(http://bit.ly/10Mi8h). This concept can be extended to any type of digital data implemented with an USB or Ethernet data stream. I believe single fiber strands are actually cheaper than running CAT-5 cable over extended distances and it allows for the use a digitized video stream directly form the source. Listed below are MPEG-2 based analogue-video/USB converters. USB can only go 5m and analogue -to - Ethernet cameras are much more expensive. I have looked long and hard for a solution and believe this is the most efficient way of concentrating multiple digitized camera streams over any distance at a single source using off the shelf mass produced equipment. Instead of using fiber the Ronja FSO design can be deployed which will be a more cost effective solution over longer distances. FpGa using JPEG IP cores are also viable but would be more costly unless mass produced. The BT878 is a single integrated chip with audio and video for which Linux drivers are available.
The BT878 outputs its data in PCI format, thus an FpGa IP would be used to convert the PCI signal to USB or Ethernet. The IP core could be anything over $10000. Startech has probably done this using a custom ASIC device, as FpGa would be to costly for mass production. Reverse engineering the PCB would be trivial to do, the hardware is the easy part, the software IP cores is the issue.
* http://www.newegg.com/Product/Product.aspx?Item=N82E16882203057 Linux based USB mpeg-4 capture * Startech SVID2USB2 uses BT878 chipset and available from http://www.newegg.com * http://www.startech.com/item/SVID2USB2-USB-2-Video-Capture-Cable.aspx
[edit] Design flaws in Ronja
Ronja uses discreet analogue components instead of the commercial IC TransimpedenceAmp(http://bit.ly/10Mi8h) TZA3033 and TZA3034. A photo PIN diode receives the modulated light pulses and converts it to a current, converted by the TZA3033(http://www.nalanda.nitc.ac.in/industry/datasheets/philips/Sec09/an98082.pdf) into millivolts and amplified to PECL voltage levels by the TZA3034. The Ronja LED transmitter though is an innovative design. With LED based FSO systems the transmission side of things is where the difficulty lies because of the limited bandwidth compared to laser systems. With Laser FSO atmospheric conditions becomes the problem resulting in unacceptable transmission errors which can only be corrected with Reed-Solomon on FpGa(http://bit.ly/13Zlxl) and OpticPatents(http://bit.ly/n3CTH) methods.
Replace the analogue receiver side of RONJA http://ronja.twibright.com/mesh.php with the SONET 155Mbit/s TZA3033 and TZA3034 design as per http://www.digchip.com/datasheets/download_datasheet.php?id=1009741&part-number=TZA3033T. The datasheet clearly shows how the TZA3033 interfaces with the TZA3034. University of Chicago based a design for the CERN project on this as per TransimpedenceAmp(http://bit.ly/10Mi8h). The TZA3034 post amplification stage, http://en.wikipedia.org/wiki/PECL signal interfaces with the RONJA Ethernet Manchestor encoding logic. Using the the TZA3033 considerably simplifies the design because no manual calibration is needed making mass production a cost effective option. Reverse engineer a normal 15Mbit/sSONET fiber(cable) optic switch to get the PCB layout and Gerber files for the photodiode(light receiver), transimpedance(TZA3033) and amplifier to PECL logic(TZA3034) stages.
The TZA3034, http://en.wikipedia.org/wiki/PECL (emitter coupled logic stage) interfaces with the RONJA Manchestor encoding stage, reducing the entire RONJA design to one of simple boolean logic instead of complex analogue integration which it is at the moment. The TZA3033 allows a bit rate of 155Mbit/s much more than the 10Mbit/s of RONJA. Thus our attention can focus on switching the Lumileds HPWT-BD00 at 20Mbit/s over RONJA or using 200Mbit/s high frequency rate LEDS like RC-LED(resonant cavity LED) coupled with the Terabeam patent to increase optical power. See the http://www.firecomms.com/tech-RCLED.html entry under FiberSwitches on using RC-LED over plastic instead of copper. During night time or where there isn't much atmospheric interference or over short distances lasers can thus be used to achieve 155Mbit/s. By daisy-chaining multiple of these cheap laser based optic FSO stages(50meters) a very high speed back bone can be created over extended distances that might be stable enough for certain periods(100% up-time not guaranteed with a laser but is with a LED) if Reed-SolomonFpGa error correction is used. With FSO(laser and LED) as with a fiber switch, unlimited repeater nodes can be created.
[edit] Indoor RONJA
Remove the loupes lenses and place the Rx against the roof with perhaps a IR filter. Use a wide-angle Tx Lumileds HPWT-BD00 without any focusing lenses for max beam dispersion inside a room.
[edit] Mixing 650nm and 880nm
Place two Ronja's adjacent to each other with the one using 650nm(red) and the other 880nm. Use bandpass filters at Tx and Rx.
[edit] Hacking FSO designs
There are many single chip FSO solutions that implements the entire optical,electrical,Ethernet stage one either a single or two chips such as http://www.micrel.com/_PDF/mic3001.pdf. Contact them to find out which companies have released commercial products in FSO using their chipsets then reverse engineer those boards releasing the PCB, Gerber files on the Internet. Get hold of a commercial FSO system such a from http://www.pavdata.com that uses a LED to get hold of the hardware interface stages which is where all the complexity lies. With laser based systems the issue is software. http://www.pavdata.com/PAVLight%20155%20datasheet_0608_IntLR.pdf uses a 910nm laser diode. http://www.pwcomms.co.uk Caution is needed to protect eyes. Hack the design and publish on the Internet.
[edit] Increase receiver size
OpticalManufacturers makes any size Loupes (loop) lens. Use a solar cooker parabolic reflector: A large parabolic or Fresnel focusing lens collects a huge area of the incoming LED Ronja signal, extending the range of a Ronja. With FSO hundreds of decentralized nodes are created for commercial Internet distribution. Not all users would need a bulky parabolic setup, the usual Loupes design for 1.4km would be enough.
[edit] Fresnel lenses
http://www.fresneltech.com/pdf/FresnelLenses.pdf Nearly all of these fresnels are designed to have their grooved side facing the longer conjugate (ie the distant station) with the flat size inside the optical enclosure facing the neaby source or photodiode. [Author's note, May 2005: A 40 cm (16 inch) square fresnel, for instance, with a focal length approximately the same as its diameter, is available from the 3DLens Company in Taiwan for US$28.50. Refer their web catalogue ]: http://www.3dlens.com
* http://www.fresneltech.com/pdf/FresnelLenses.pdf * http://www.bikudo.com/products.do?tmp=1&show_save=yes&keyword=Fresnel+Lens&search_method=2&sa=1
Fresnel lenses of 300 x 300mm to focus Ronja LED light on PIN diode.
* http://www.bikudo.com/product_search/details/45452/fresnel_lens_mold.html
Fresnel lens mold for custom production of fresnel lenses.
Wide beam technology overcomes building movement problems http://mybroadband.co.za/vb/showthread.php?t=174276&page=2
[edit] DWDM or CWDM over fiber hack
DWDM (Dense wave division multiplexing), http://en.wikipedia.org/wiki/CWDM (Course wave division multiplexing). Install a single mode fiber optic cable between users. Combine upto 40 multiple wavelengths as per http://www.patentstorm.us/patents/6868236/fulltext.html (Terabeam). Receive the data using an optical bandpass filter - http://www.edmundoptics.com/onlinecatalog/browse.cfm?categoryid=10 , http://www.optical-coating.com/cata_f.html , http://www.oceanoptics.com/products/optics.asp at each node spacing down the length of the cable with a TransimpedenceAmp on 660nm, 700nm, 800nm, 900nm....1550nm(40 total). Each node uses a passive optical tap (http://www.diconfiberoptics.com/products/prd_passive.php) connected to a bandpass filter , which will reduce the range of the laser diode signal. A laser diode pulses the data instead of a LED. Reverse engineer an UTP-to-Optical two-port fiber switch such as :.........ENTHER HERE........ They usually come in in 1550nm and 980nm only, but any wavelength from 600nm can be used. The main drawback is the cost of commercial bandpass filters, more research is needed in hacking our own optical bandpass technology, there probably are a few patents driving up the costs to the present $100 per filter. Multiple optical repeater routers can be installed down this single optical fiber decreasing the effective cost of a fiber installation.
[edit] Links
http://linkinghub.elsevier.com/retrieve/pii/S0030401801015917 fresnel lenses for IR reception
[edit] 100 meg is possible with LED
Lasers can't be used because of atmospheric effects. The switching frequency of leds have increased to the point where 100meg is possible.
* http://translate.google.com/translate?prev=hp&hl=en&u=http%3A%2F%2Fronja.vyrobce.cz%2F&sl=cs&tl=en * http://ronja.vyrobce.cz/
[edit] 10Gig FSO link
Discovery Semiconductor http://www.chipsat.com/ demonstrates free-space optics link with AFRL Schafer, SAIC participate in 10-Gbit link using RZ-DPSK modulation http://www.commsdesign.com/design_center/opticalnetworking/news/showArticle.jhtml;jsessionid=YTTHJQ1QAPUJ4QSNDLRSKH0CJUNN2JVN?articleID=210605297
The article concludes that RZ-DPSK works best to overcome scintillation effects in laser transmission systems.
[edit] Reverse engineer commercial 1Gps FSO
Reverse engineer -BoMarc the commercial units such as Terabeam(OpticPatents) and release the designs somewhere on the internet. It will cost around $3000 to reverse engineer the PCB using in http://www.armisteadtechnologies.com/reverse-engineering.shtml. The patents which might cover the design isn't a problem , use a FrontingCompany since we are not in it for commercial gain. The software involves Reed-Solomon for which ElectrConsultant are available. Be careful though when working with a consulting engineer, many of them are great fans of opensource Linux just as long as their hardware and DSP knowledge doesn't get opensourced. Some will request that a NDA be signed and prefer to work with a company who wants to commercialize a product. Kestrel engineering - SoftwareDefinedRadio will be able to help with opensource hardware projects. We in South-Africa are not trying to commercialize it but save this country from an abyss of crime via NetworkCentricWarfare. To study the transimpedence pre and post amplifiers of a normal fiber based switch reverse engineer it.
[edit] FPGA based error correction of laser optic networks
Reed-Solomon encoding, decoding FPGA from http://scratchpad.wikia.com/wiki/FpGa#Asics_encoder-decoder_cores are used to correct FSO laser 2.5Gig transmission errors. It is not needed though with LED 10-100meg bit rates.
[edit] Factors influencing FSO
http://www.seas.harvard.edu/hbbcl/fsoc.html ,,, http://www.free-space-optics.org/fso_technology.php
Fog: Fog is vapor composed of water droplets, which are only a few hundred microns in diameter but can modify light characteristics or completely hinder the passage of light through a combination of absorption, scattering, and reflection. This can lead to a decrease in the power density of the transmitted beam, decreasing the effective distance of a free space optical link. Fog substantially attenuates visible radiation, and it has a similar affect on the near-infrared wavelengths that are employed in laser communications. Similar to the case of rain attenuation with RF wireless, fog attenuation is not a “show-stopper” for optical wireless, because the optical link can be engineered such that, for a large fraction of the time, an acceptable power will be received even in the presence of heavy fog. Laser communication systems can be enhanced to yield even greater availabilities by combining them with RF systems.
Scintillation: Scintillation is the temporaland spatial variation in light intensity caused by atmospheric turbulence. Such turbulence is caused by wind and temperature gradients that create pockets of air with rapidly varying densities and, therefore, fast-changing indices of optical reflection. These air pockets act like lenses with time-varying properties and can lead to sharp increases in the bit-error-rates of free space optical communication systems, particularly in the presence of direct sunlight. Performance of many laser communications systems is adversely affected by scintillation on bright sunny days. Through a large aperture receiver, widely spaced transmitters, finely tuned receive filtering, and automatic gain control, downtime due to scintillation can be avoided. Use ReedSolomon encoding to minimize the error rate.
Beam Wander: Beam wander arises when turbulent wind current (eddies) larger than the diameter of the transmitted optical beam cause a slow, but significant, displacement of the transmitted beam. Beam wander may also be the result of seismic activity that causes a relative displacement between the position of the transmitting laser and the receiving photodetector Terrestrial Laser Communications Challenges
Physical Obstructions: Laser communications systems that employ multiple, spatially diverse transmitters and large receive optics will eliminate interference concerns from objects such as birds.
Pointing Stability:Pointing stability in commercial laser communications systems is achieved by one of two methods. The simpler, less costly method is to widen the beam divergence so that if either end of the link moves the receiver will still be within the beam. See http://thorlabs.com/navigation.cfm?Guide_ID=10 The second method is to employ a beam tracking system - FsoPatents. While more costly, such systems allow for a tighter beam to be transmitted allowing for higher security and longer distance transmissions.
[edit] Fiber like repeater functionality
The Ronja implements IEEE 802.3 Ethernet and like a fiber node switch allows nearly unlimited repeater links of Ronja's back-to-back before latency becomes an issue depending on the switch type either cut through or store forward - FsoRepeater
[edit] Meshing with FSO
http://ronja.twibright.com/mesh.php Meshing up homes, offices, and buildings: Multiple installations in an area form a mesh. The mesh can exploit multiple redundancies in paths, which can be managed by OSPF and/or BGP protocols. Ownership of the network may be distributed among the participant and still reliable operation can be ensured by employing these protocols.
Meshing with mixed technologies :Ronja can be easily mixed with other network technologies without constraints. Therefore meshes mixing Ronja with ISM band radio wireless devices is for example possible, either as a solution for transition from radio-based mesh to an optical mesh, or as a newly developed solution.
[edit] Costs
Outsource the construction to a person paid around R4000/month to hand solder multiple units.
[edit] Oreillynet
http://www.oreillynet.com/etel/blog/2007/02/ronja_at_10_mbps_the_next_stag.html "...The promise of Ronja is mesh technology that can deliver 10 Mbps and can be built by an amateur in his or her own home for $100 per unit. The specs are all open-source. Where costs or regulations delay the stringing of cable or fiber, this technology could quickly bring neighborhoods into the twenty-first century in terms of bandwidth and universal service. Applications such as interactive video teleconferencing and remote application access with large remote data storage become immediate possibilities. The Ronja team currently counts 146 registered installations in at least 9 countries. The largest installation is in Prague, where the http://czfree.net/home/index.php community network has 29 links built with Ronja transmitters - http://ronja.twibright.com/czfree.php. A glance at the credits and mailing list archives - http://lists.pointless.net/pipermail/ronja/ reveals a passionate community of amateurs sharing insights and images concerning what parts to order and how to solder, clamp, and install Ronjas. The team is hoping in the near future to extend the range to 3 km and the speed to 100 Mbps. Eventually, with a laser costing $2, they hope to reach 1 Gbps......"
[edit] Beam alignment
http://sections.menzonet.org/metropolis/guides/ The picture shows what the transmitter beam looks like from 260m distance (130mm diameter transmitter lens) and the big image is the 90mm transmitter in operation from close, out-of-axis view. Now wait until dark (or at least twilight) and aim the transmitter onto some distant house. Loosen the focus (with #7 wrench) and focus until you see a bright spot. Then tighten. Move the transmitter onto the retroreflector and play with both M8 bolts (two #13 wrenches) and the focus until you get maximum brightness on the retroreflector. Then carefully tighten and watch the spot to be sure you have not drifted during tightening. Measure the received signal strength shortly before you leave the roof and write it down and put it into a safe place for future reference to be able to say if something went wrong.
Ask the other party to aim their transmitter. Then connect the DC voltmeter on 200mV range to the measuring port of the receiver. Then put the focus of the receiver into middle position and fuss around with the receiver until you see some numbers on the meter. Now play with the focus and aiming until you get maximum count. Now freeze down the setting by tightening the nuts down, again watch the meter not to decrease the count (which would mean corrupting the optimum aiming).
[edit] Lasers and eye safety
Use 1550nm for eye safe laser transmission or a pulse controlled Ronja system on 880nm. The Tx sends out brief eye safe pulses and waits for a response. Any blocking of the laser will result in Tx sending out only brief pulses until the other Tx indicates no obstruction in the path.
[edit] Laser instead of LED
http://www.pointless.net/pipermail/ronja/2003-August/000709.html At 04:24 PM 8/7/2003 +0000, Patrick Deelman wrote:
>So my question is whether someone is still working on a 100mbit version. I >sure would be intrested in that :)
<AOL> Me too! </AOL>
>Ow and by the way. Why is that radio amateurs are reaching distances from >100 miles (160km) and more (okay i know they
Well, there are a couple of things working against Ronja...
The lower the wavelength, the further the signal will go. Light is a really, really, really high wavelength. (terahertz)
The corollary is that the higher the wavelength, the more information carrying capacity you have. Which means that at lower wavelengths you need wide swaths of spectrum to get a decent amount of information across. High wavelengths such as light can transmit a staggering amount of information in a very narrow frequency range. Ronja is woefully under using the information carrying capacity of light. (think fiber optics)
Finally, radio frequencies are regulated. Light currently isn't. So, to answer your question.. yes, radio amateurs do go 100s of kms and more. They use up huge chunks of frequencies to do so and ultimately their fastest speeds are only a match for the crudest optical setups. More refined optical links will far outpace them. Now, to address distance in Ronja. First and foremost the enemy of distance is aiming this sucker. As you extend the distance, even minute changes on the transmitter end translate into wide arcs at the receiving end. Please realize no mount is 100% solid - even the most solid building sways slightly in the wind, mounts change with temperature, and more. What you need to counteract this all is a precise aiming system. Complex and costly.
Ronja could switch to lasers, it adds cost but gets more photons over the distance. The question with lasers is eye safety - as you crank the power you can cause real damage to people who inadvertently look in the laser's direction - or worse yet point a pair of binoculars in the laser's direction. Every laser beam spreads over distance as a result of focusing and scattering, and that circle can be many meters wide at the receiving end - so it is possible that someone with a window near the receiver could be hurt. To address someone sticking their face in the transmitter lens, there are systems that quickly shut off the laser if contact is lost and does only very quick and eye safe pulses to regain contact, but that once again adds costs. Ronja's current LEDs are safer - they're so bright you turn away and because they're not coherent laser light they won't seriously hurt your eyes for that brief exposure. But that lack of coherency makes them less likely to cross the distance - the non-coherent light, no matter how well it is focused, won't stay together as well as a laser. It's also difficult to find high powered LEDs that switch fast enough to sustain 100Mb.
Finally, there's the curve of the planet. Light goes in a straight line. Lower frequency radio waves can curve around the horizon. This means for a significant distance, you need to get both transmit and receive way high up on a tower to see over the horizon. Microwaves (GHz radio waves) have the same limitations. So, in summary, add precise and constantly adjustable aiming to Ronja, (and the logic to control that aiming) crank up the brightness, maybe add a encoding method that is handles noise better, maybe use lasers, and you should be able to get significant distances. Of course, all this makes Ronja harder and more expensive to build.
For me, the perfect Ronja is 100Mb, self-aiming over shorter distances (0.5km to 2km), eye safe, is a circuit that works on a board, and has a auto negotiating twisted pair port. (ie. something that only needs course aiming and doesn't require going out and constantly re-aiming it) With that my friends and I can set up a reliable, high speed, mesh to share bandwidth.
ps. The perfect, perfect Ronja is one that has both the Ethernet TX/RX, Optical TX/RX, and a couple of auxiliary lines into fed into a FPGA. This would give us a programmable Ronja - people could program the FPGA to do whatever they'd like (different link encoding, encryption, autoneg, speed negotiation on the optical link, steering logic, whatever) using a single hardware design. Don't know if that's feasible - I'm a software/logic guy, not a hardware guy. If the hardware is standard, it would be feasible to maybe even construct and sell a kit as a business...
[edit] Spectrum scarcity
Instead of each home polluting the spectrum with a WiMax CPE, ten houses forms a hotspot via FSO and only one CPE is shared between them. Vodacom and the NaspersProblem are lying about spectrum availability. There is more than enough spectrum if FSO hotspots are created first. FSO transmitting 1Gps over 2km reduces the need for fiber. Vodacom's specific problem is that they can't request residences to willing house their FSO transmitters and then expect that house to be defrauded by Vodacom to the tune of R5/min for a phone call. We should specifically sell commercial internet via such FSO links which would drive the 2.4ghz and 5.8ghz spectrum polluting Wisps out of business. With 2.4ghz and 5.8ghz spectrum freed up WiMax and MeshNetworking should be used exclusively to connect FSO networks at distances exceeding 4km. We would be able to cover Pretoria and Gauteng in a broadband network in a matter of months.
[edit] Icasa issues
[QUOTE=dominic;2892039]once more into the breech ...
use of FSO does not require a spectrum licence but will
- require an ECNS licence for the hardware (unless it falls within an exemption as a private network)
- require an ECS licence for the internet services provided over it (unless it is non-profit or resale) [/QUOTE]
[url]http://scratchpad.wikia.com/wiki/FreeSpaceOptics[/url] using the Ronja LED system won't be licensed by Icasa for obvious reasons. Thus the answer is that Icasa will confiscate any FSO (LED based costs aroun R1500 for 1.4km distance) used to provide a commercial service without a license. In practice this won't be so easy because FSO can be "sprinkled" all over the place because there are no Fresnel zone issues. The laser FSO systems are to expensive at present unless we hack the patents.
[QUOTE=dominic;2892039]once more into the breech ... resale of internet via wifi is licence exempt (the licence exemption must be registered with ICASA at no cost) [/QUOTE] But since the Ronja system won't be approved for commercial usage by Icasa, they will still confiscate the gear if they can, if used for commercial purposes. Wi-Fi is extremely vulnerable for confiscation because of the need for high towers. Icasa's main weapon of terror seems to be that they can destroy a tower with a cutting torch if you keep on placing back Wi-Fi gear, but Vodacom won't be allowed by the courts to break down your house if you keep on replacing a FSO node on your roof.
Build a FSO backbone transition node with 10 adjacent homes via fixed cabling, install a FSO on each roof and then provide a commercial service. Icasa can only target one home at a time, giving the others time to remove theirs, the constitution forbids Icasa from sealing of an entire street with police helicopters and armed swat team members , with hot coffee and buns provided by Vodacom. The costs of these Ronja systems can be reduced by integrating the electronic logic (Manchester encoding) on a single [url]http://scratchpad.wikia.com/wiki/FpGa[/url], instead of the present discreet component solution from Ronja. They seem to be using this due to the GPL license limitations. In South Africa we don't bother with GPL, we don't have a working legal system anymore.
[QUOTE=dominic;2892039] but, like the bunny i am not convinced of the merits of FSO as an access network [/QUOTE]
With the latest advances in LED switching a 100Meg using LEDS alone is I believe now possible. 100meg over 1.2km with unlimited repeater nodes(like a fiber switch) makes FSO a Wi-Fi killer. Even the distace with a LED system can be extended to 3km if very large Jewelers Loupes is used, but off-the shelf Loupes are massed produced only to a certain size. Larger must be specially produced but there is a minimum quantity of around 1000. The max size and distortions it introduces should be researched further. The manual manufacture of larger lenses should be explored, even Fresnel lenses is an option.
[edit] opensource led drivers
* http://en.wikipedia.org/wiki/RONJA * http://ronja.twibright.com/transmitter/building_pcb.php * http://en.wikipedia.org/?title=Free_machine
[edit] Theory behind FSO
* http://www.kultt.co.za/awiki/mediawiki/line_coding.html Line coding * Manchester encoding used in 802.3 Ethernet
[edit] Links
* TransimpedenceAmp , AmazonForum Books on optical networking * RonjaChips - chips used in Ronja * OpticEthernetBridge * OpticalManufacturers The larger the focusing lens the longer the range * ElectrConsultant - Consultants on FSO designs * OpticPatents - Noah, Aoptix, Lightepointe, Canon, Terabeam, Oplink, Alcatel, Trex, * PaTents * FsoPublication * IrDa * AlibabaOptics * LensFocusingIdeas * TempNotesFso
[edit] Links
* http://www.sctimes.com/article/20090116/NEWS01/101160018&referrer=FRONTPAGECAROUSEL * http://en.wikipedia.org/wiki/Free_Space_Optics * http://www.free-space-optics.org/fso_comparisons.php 1500nm * http://www.ok2kkw.com/next/qw2008_33km.htm 32km link * http://www.pointless.net/pipermail/ronja/2003-August/000709.html * http://linas.org/mirrors/atrey.karlin.mff.cuni.cz/2002.01.03/~clock/twibright/ronja/electric.html * http://coldstonelabs.org/doku.php?id=ronja * http://www.oreillynet.com/etel/blog/2007/02/ronja_at_10_mbps_the_next_stag.html * http://hackaday.com/2005/06/13/ronja-optical-data-link/ * http://www.alibaba.com/product-gs/209982219/plate_lens.html plate lens * http://www.thorlabs.com/ Reverse engineer photo detectors via BoMarc. * http://www.diconfiberoptics.com/products/?prod=0090&menu=ftr&sub=0
[edit] Commercial
* http://www.optica.ru/ * http://www.freespaceoptics.com/index.php?page=learnmore * http://www.fsona.com/
http://www.futurend.co.uk/free-space-optics/index.php GeoDesy FSO is part of the GeoDesy Group GeoDeys's patented Automatic Inbound Power Control guarantees
* http://www.usaccess-llc.com/mrv.html?gclid=CNfWuZ_4i5gCFQ5NQgodkR-5ew * http://www.usa.canon.com/html/industrial_canobeam/canobeam/index.html * http://www.lightpointe.com/products/fs_g.cfm * http://www.fsoalliance.com * http://www.pavdata.com * http://shop.ebay.com/merchant/fsomn_W0QQ_nkwZQQ_armrsZ1QQ_fromZQQ_mdoZ * http://www.redlinesa.co.za/freespaceoptics.html Commercial system * http://www.futurend.co.uk/free-space-optics/index.php * http://www.americantechsupply.com/mrv_free_space_optics.htm * http://www.terabeam.com/ * http://laseritc.com * http://www.moctkom.ru/indexeng.htm Artolink * http://www.mrv.com/ * Lightpointe
