Free space optics

TableOfContents

manchester
The PHY layer LAN8720 chip decodes the Manchester encoded data from the Ethernet port into a PECL serial data stream. A series of 74HC245 buffer chips drives a LED with the modulation scheme of OOK(On Off keying). The transmitted light is amplified with joulers loupes, fresnel lenses and received with similar construction. At the receive point a PIN diode(Infineon SFH203) emits a weak current for each high light pulse, which in turn is amplified into millivolts with a TIA (trans impedance amplifier)(TZA3032). The millivolts are converted into a 5v PECL stream with post amplifier TZA3034, which is read by the receiving PHY LAN8720. The PHY adds Manchester encoding, interfacing with the Ethernet port of the receiver.

We need not concern ourselves with the details of Manchester, as this is handled by the PHY layer chip. Our concern is the LED pulse transmission with OOK. http://ronja.twibright.com encodes and decodes Manchester with discreet IC's before OOK transmission with the HPWT-BD00 LED, which is inefficient. Pointing this out on their mailing list will lead to banning as they are reselling PCBs, assembly takes around a week. Ronja's 74HCT245 discreet IC ciruitry for OOK transmission is about all we need for a mass producible solution. FiberSwitches, plastic over fiber resonant cavity led is a 200Meg solution, but will need larger fresnel objective and loupes field lens for long range transmission. https://www.youtube.com/watch?v=XKtxxZ327UM 2min tutorial https://www.youtube.com/watch?v=8BhjXqw9MqI 3 of 13 tutorial he importance of synchronized clocks and using Manchester coding to send clock and data https://hardwaresecrets.com/connecting-two-pcs-using-a-usb-usb-cable/ usb to usb with bridging chip, use pre and post amplifier(TZA3032) for 5v PECL stream.

Koruza

 * https://github.com/IRNAS/KORUZA wireless optical project with subs https://github.com/IRNAS/koruza-compute-module
 * http://forum.irnas.eu/viewtopic.php?f=7&t=3
 * http://instructions.koruza.net/
 * https://github.com/IRNAS lots of projects
 * http://irnas.eu/UsefulSource/koruza/koruza1.xml
 * https://www.youtube.com/watch?v=yq6hBNCzH10 Developer video, https://www.youtube.com/watch?v=Ke-LlxBfutc
 * Uses https://en.wikipedia.org/wiki/Small_form-factor_pluggable_transceiver SFP transceiver . video on what a SFP is and how to use it
 * http://datainterfaces.com/bidirectional-wdm.aspx bidi sc with ddm, sfp module
 * https://www.perle.com/products/sfp-optical-transceiver.shtml#sfp-simplex-bidi
 * http://instructions.koruza.net/instructions/bom.html uses http://www.tp-link.com/us/products/details/cat-43_MC220L.html The MC220L is a media converter designed to convert 1000BASE-SX/LX/LH fiber to 1000Base-T copper media or vice versa. Designed under IEEE802.3ab 1000Base-T and IEEE802.3z 1000Base-SX/LX/LH standards, the MC220L is designed for use with multi-mode/single-mode fiber cable utilizing the SC/LC-Type connector. It works at 850nm on both transmitting and receiving data when adopting multi-mode fiber and 1310nm on both transmitting and receiving data when adopting single-mode. Other features of this module include the ability to be used as a stand alone device (no chassis required) or with TP-LINK?s 19??system chassis, Auto MDI/MDI-X for TX port, and front panel status LEDs. The MC220L will transmit at extended fiber optic distances utilizing multi-mode fiber up to 0.55 kilometers and 10 kilometers when utilizing single-mode fiber.

RONJA Links

 * http://ronja.twibright.com 10Meg link under GPL license.
 * FiberSwitches
 * Plastic optical fiber The datasheets don't seem to allow any power above 100ma peak(1us), Ronja forces over 1amp peak through a non rcled, but can't switch above 10meg. Rcled switches at 250meg - 1.25Gig. Focus the RCled through a jewlers loupes and Fresnel lens combination, this should achieve at least 50meters free space. See the FSO category of fresnel, loupes lens combination on laser transmission. Latency on each FSO,rcled or ronja node is less than 1us, which means that daisy chaining a node every 50m for 50km will make for 10ms latency in total.

With http://ronja.twibright.com unlimited repeater nodes are possible, it functions as a layer-3 optical switch. The design though is crippled because it uses discrete op-amps for the pre-amplification stage, instead of IC pre-amps/post-amps as found in copper to fiber ethernet range extenders. Pre-amp IC contains hundred of transistors, enabling 100Meg and above signal detection, which can't be done with analogue designs. Resonant cavity LEDS (FiberLed) and http://sasecurity.wikia.com/wiki/OpticPatents#Margaritis.2C_Georgios enables 100Meg to 1gig design. Scintillation only affects laser above 50m, allowing non-complex laser designs multiple bridges from residence to residence. FSO, without PID control and incorporating the OpticPatents RcServos multiple pin diode idea.

SDH/SONET STM1/OC3 optical receiver 100meg is available and a very stable 10meg PECL stream to interface to the Ethernet PHY layer IC.
 * http://ronja.twibright.com 1.2km 10Meg link under GPL license.
 * http://www.ohwr.org/projects/white-rabbit  CERN white rabbit ethernet project, use it to replace the transistors in Ronja with pre-amp/post-amp PECL,  PHY ,MAC layer IC chipset. It is difficult to comprehend why Ronja would want to force users to solder for 70hours. This is why the ronja project is going nowhere, people don't have time to hack. Only the LED driver stage would need to be retained from the original ronja design. With the CERN white rabbit IC pre-post amps like the TZA3300
 * RonjaChips, RonjaFpga, GmaxwellRonja
 * FiberSwitches or FiberSwitches Plastic optical switching at 200Mbit/s using RCLED that can switch at 400Mbit/s. The firecom ICs have the pre-amplification and post-amplifiation stage built into a single IC. See TransimpedenceAmp for reference.
 * EthernetTransceiver Corrects baseline wander, converts Ethernet PHY to Fiber compatible
 * Lxt971Aphy Ethernet transceiver stage with fiber output with Linux drivers..
 * OpticPosts Usenet posts on electrical light conversion.
 * RonjaChips, RonjaFpga, GmaxwellRonja, OpticEthernetBridge, OpticalManufacturers
 * SerDes, AmazonForum, ElectrConsultant, OpticPatents Design details , part numbers and linear positioners.
 * PaTents, FsoPublication,IrDa, InfraRedLeds, SerialDrivenInfraRed
 * TransimpedenceAmp Google ROSA, TIA, PIN
 * CharlesFalconeSerialRoutines, FiberLed, OpticalPapers , GizmagNotes
 * LowCostLongRange, DslamRonja, RcServos Linear positioner for use in Bowhunting and FSO positioning
 * MicrelLaser,
 * HamamatsuPhoton FSO modules for pulsed light detection through free air.
 * Sony Optics TIA, laser drivers
 * FiberEthernetModule

TOSA transmitter optical sub assembly and ROSA Receiver optical sub assembly

http://www.google.co.za/search?aq=f&gcx=c&sourceid=chrome&ie=UTF-8&q=tosa+#sclient=psy-ab&hl=en&source=hp&q=tosa+optics&pbx=1&oq=tosa+optics&aq=f&aqi=g-e1g1g-v1g-sv1&aql=1&gs_sm=e&gs_upl=14895l16251l0l16385l7l6l0l0l0l0l406l1680l2-2.2.1l5l0&bav=on.2,or.r_gc.r_pw.,cf.osb&fp=3829c9d23f3c3799&biw=1366&bih=643

Roithner fiber LED light to fiber coupling
Roithner lasers from InfraRedLeds has a glass spiral mould that takes diffused light such as LED and focuses it into a optic fiber cable. Hack the http://sasecurity.wikia.com/wiki/FiberSwitches firecomms, replacing the polymerr plastic with a fiber cable.

Fiber link
To understand how a FSO link is created lets first create a dual mode optical link between two generic CAT-5(copper) based Ethernet switches using fiber. Ethernet standard specifies three different voltage levels output at the PHY layer. Logic circuitry only understands two voltage levels +5 and 0.

EthernetTransceiver (OpticPatents) which decodes Ethernet at the MAC and PHY layer,converting three volt levels to a binary stream (PECL levels) for a fiber interface, modulated with NRZ.
 * From the PC Ethernet port a CAT-5 cable(eight lines) connects with the Intel Lxt971Aphy, a Dual-Speed Fast

http://www.optekinc.com/datasheets/OPF2418.pdf. (See OpticPosts and OpticalManufacturers). The OPF2418 converts the pulses back to copper voltage PECL output.
 * NRZ binary stream drives a laser or diode signal(on/off),sending light pulses at 10Meg over the fiber to a TIA(TransimpedenceAmp) such as


 * The PECL from the OPF2418 interfaces with the Intel Lxt971Aphy chip at the other end, which demodulates the NRZ stream by sampling the signal in the middle of the bit and converts it back to three volt levels to interface with the Ethernet switch at the PHY layer.


 * With FSO the fiber link is replaced with the free air medium, FresnelLens, Collimator etc.

Lxt971Aphy automatically switches between either coding format. There was no need for Ronja to use discreet components for something a single Lxt971Aphy implements. * : Realtek RTL8213 ethernet chip
 * Take a UTP-FIBER converter commercially available for around $100 and reverse engineer the design for a PCB via BoMarc. Get hold of a http://en.wikipedia.org/wiki/Fast_Ethernet#100BASE-SX with LED driver for a FSO design. HackPatents explains why it isn't unethical to reverse engineer circuits, copy them and then sell products bases on such designs.
 * Tx: switch cat5 -> Lxt971Aphy ->Laser/LED driver -> LED or laser diode -> Fiber
 * Rx: Fiber -> TIA -> Lxt971Aphy -> CAT5 -> switch. At 10Meg Manchester encoding is used, at 100Meg NRZI. The
 * http://www.alldatasheet.com/view.jsp?Searchword=ML6652 ML6652 fiber and copper media converter
 * http://www.alldatasheet.com/view.jsp?Searchword=DM9301 DM9301 fiber and copper media converter
 * http://www.alldatasheet.com/view.jsp?Searchword=IP113ALF IP113ALF 100BASE-TX/FX media converter
 * LED driver is MAX3263
 * Philips SA5211 TransimpedenceAmp, http://www.chipdocs.com/datasheets/datasheet-pdf/Philips-Semiconductors/SA5211.html , post amplification stage http://www.datasheetcatalog.org/datasheet/philips/SA5217_2.pdf

TIA ROSA has a RSSI voltage output proportional to the light intensity received, monitor this with A/D..

I am working on a 100 mbps design based on ML6652 and LED driver is > MAX3263 and Reciver uses philips SA5211 as transimpedance Amplifier > .Will be releasing first shematics in a day or two. I am planning to > use VCSEL laser to begin with.

Irda vs. Ronja
Irda implements a protocol stack to compensate for the Tx flash-over to Rx due to the close proximity of Tx and Rx on the same chip. With Ronja we have a physical barrier between Tx and Rx so that only Ethernet can be implemented.

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.

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.

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.

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.

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

Wireless Data using visible Led light
http://www.lvxsystem.com/ http://www.google.com/patents/US20120230696?dq=John+Pederson&hl=en&sa=X&ei=39SAUf7TPKir7AbesoHQBA&ved=0CEQQ6AEwBA
 * http://www.inhabitat.com/2010/03/11/scientists-find-way-to-turn-led-lights-into-wireless-internet-source/
 * http://inhabitat.com/company-finds-way-for-led-ceiling-lights-to-provide-internet-access/ Commercialized by
 * Hack the lvx patent here by John Pederson or John C. Pederson
 * http://www.sciencedaily.com/releases/2010/03/100309151503.htm The concept is to filter out the blue spectrum allowing 250meg switching.
 * Get the Ted talks on white LEDS forming a network.
 * http://usatoday30.usatoday.com/tech/news/2011-01-07-LED-LIGHTS_N.htm
 * Inventors
 * Paul R. Brown
 * James LeClaire
 * Brent Mikkelsen
 * John C. Pederson
 * Timothy J. Vogt
 * James Zimmerman

Fiber link
To understand how a FSO link is created lets first create a dual mode optical link between two generic CAT-5(copper) based Ethernet switches using fiber. A UTP/fiber converter links two switches over dual mode fiber. FSO link should be seen as a dual mode fiber converter, but instead of using a fiber cable the free air a medium is used. Ethernet standard specifies three different voltage levels output at the PHY layer. Logic circuitry only understands two voltage levels +5 and 0.

EthernetTransceiver (OpticPatents) which decodes Ethernet at the MAC and PHY layer,converting three volt levels to a binary stream (PECL levels) for a fiber interface, modulated with NRZI.
 * From the PC Ethernet port a CAT-5 cable(eight lines) connects with the Intel Lxt971Aphy, a Dual-Speed Fast

http://www.optekinc.com/datasheets/OPF2418.pdf. (See OpticPosts and OpticalManufacturers). The OPF2418 converts the pulses back to copper voltage PECL output.
 * NRZ binary stream drives a laser or diode signal(on/off),sending light pulses at 10Meg over the fiber to a TIA(TransimpedenceAmp) such as


 * The PECL from the OPF2418 interfaces with the Intel Lxt971Aphy chip at the other end, which demodulates the NRZ stream by sampling the signal in the middle of the bit and converts it back to three volt levels to interface with the Ethernet switch at the PHY layer.


 * With FSO the fiber link is replaced with the free air medium, FresnelLens, Collimator etc.

* Tx: switch cat5 -> Lxt971Aphy ->Laser/LED driver -> LED or laser diode -> Fiber * Rx: Fiber -> TIA -> Lxt971Aphy -> CAT5 -> switch. At 10Meg Manchester encoding is used, at 100Meg NRZI. * http://www.alldatasheet.com/view.jsp?Searchword=ML6652 ML6652 fiber and copper media converter * http://www.alldatasheet.com/view.jsp?Searchword=DM9301  DM9301 fiber and copper media converter * LED driver is MAX3263 * Philips SA5211 TransimpedenceAmp, http://www.chipdocs.com/datasheets/datasheet-pdf/Philips-Semiconductors/SA5211.html , post amplification stage http://www.datasheetcatalog.org/datasheet/philips/SA5217_2.pdf * : Realtek RTL8213 ethernet chip
 * Take a UTP-FIBER converter commercially available for around $100 and reverse engineer the design for a PCB via BoMarc. Get hold of a http://en.wikipedia.org/wiki/Fast_Ethernet#100BASE-SX with LED driver for a FSO design. HackPatents explains why it isn't unethical to reverse engineer circuits, copy them and then sell products bases on such designs.

TIA ROSA has a RSSI voltage output proportional to the light intensity received, monitor this with A/D..

I am working on a 100 mbps design based on ML6652 and LED driver is > MAX3263 and Reciver uses philips SA5211 as transimpedance Amplifier > .Will be releasing first shematics in a day or two. I am planning to > use VCSEL laser to begin with.

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.

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.

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.

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. (NOTE: THIS WON'T WORK, it is a draft post)

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

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, but LED driver is MAX3263 provides the same if not better functionality.Eight HCL chips in series takes up a lot of board space, increasing the BOM. A single MAX3263 is the preferred solution, allowing mass production. 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 beyond 50meters 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. Below 50meters a highspeed link to cross a road is possible without atmospheric interference having an effect.

At night or in cloudy conditions a long range(4km p-t-p) laser link is possible without OpticPatents methods, allowing a full Internet experience during night time instead of 24hours would be acceptable in most circumstances.

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.

Replace the discrete Manchestor logic of Ronja with an EthernetTransceiver Lxt971Aphy for which there is GPL http://en.wikipedia.org/wiki/Das_U-Boot code available. IEEE 802.3 Ethernet uses Manchester at 10Meg and NRZI at 100Meg.

Using the the TZA3033 TransimpedenceAmp 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.

Das U-boot
Replace the discrete Manchestor logic of Ronja with an EthernetTransceiver Lxt971Aphy for which there is GPL http://en.wikipedia.org/wiki/Das_U-Boot code available. IEEE 802.3 Ethernet uses Manchester at 10Meg and NRZI at 100Meg.

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.

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.

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.

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.

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/Large_Fresnel_Lens.html * 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.
 * http://www.modulatedlight.org/Modulated_Light_DX/OpticalComms4Amateur79.html

Wide beam technology overcomes building movement problems http://mybroadband.co.za/vb/showthread.php?t=174276&page=2

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.

Links
http://www.google.co.za/search?hl=en&as_q=company+fresnel+lens&as_epq=optics&as_oq=size&as_eq=&num=100&lr=&as_filetype=&ft=i&as_sitesearch=&as_qdr=all&as_rights=&as_occt=any&cr=&as_nlo=&as_nhi=&safe=images

http://linkinghub.elsevier.com/retrieve/pii/S0030401801015917 fresnel lenses for IR reception

100 meg is possible with LED
Lasers can't be used because of atmospheric effects, exceeding 50m. 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/

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.

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.

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.

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.

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

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.

Costs
Outsource the construction to a person paid around R4000/month to hand solder multiple units.

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......"

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).

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.

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 :)

 Me too! 

>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...

opensource led drivers
* http://en.wikipedia.org/wiki/RONJA * http://ronja.twibright.com/transmitter/building_pcb.php * http://en.wikipedia.org/?title=Free_machine

Theory behind FSO
* http://www.kultt.co.za/awiki/mediawiki/line_coding.html Line coding * Manchester encoding used in 802.3 Ethernet

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

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

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

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.

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.

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.

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.

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/Large_Fresnel_Lens.html * 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.
 * http://www.modulatedlight.org/Modulated_Light_DX/OpticalComms4Amateur79.html

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.

Links
http://www.google.co.za/search?hl=en&as_q=company+fresnel+lens&as_epq=optics&as_oq=size&as_eq=&num=100&lr=&as_filetype=&ft=i&as_sitesearch=&as_qdr=all&as_rights=&as_occt=any&cr=&as_nlo=&as_nhi=&safe=images

http://linkinghub.elsevier.com/retrieve/pii/S0030401801015917 fresnel lenses for IR reception

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/

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.

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.

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.

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.

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

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.

Costs
Outsource the construction to a person paid around R4000/month to hand solder multiple units.

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......"

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).

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.

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 :)

 Me too! 

>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...

opensource led drivers
* http://en.wikipedia.org/wiki/RONJA * http://ronja.twibright.com/transmitter/building_pcb.php * http://en.wikipedia.org/?title=Free_machine

Theory behind FSO
* http://www.kultt.co.za/awiki/mediawiki/line_coding.html Line coding * Manchester encoding used in 802.3 Ethernet

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

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

Publications
http://www.optoiq.com/index/photonics-technologies-applications.html

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.moctkom.ru/

* 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 * http://www.moctkom.ru/ * http://www.artolink.com/

1Gps with laser pointers
visible light, won't work over large distances.
 * http://www.extremetech.com/extreme/128207-1gbps-wireless-network-made-with-red-and-green-laser-pointers
 * http://www.opticsinfobase.org/oe/abstract.cfm?uri=oe-20-9-9919