Slashdot Mirror


Alternative Frequency Wireless Ethernet Devices?

rtgree01 asks: "Do Slashdot readers know of any wireless ethernet devices that are not 900 MHz, 2.4GHz, or 5GHz. This is for a project that will be in an extremely harsh EMI area (thousands of V/m at those frequencies). Also, to make things even more impossible, line of sight is not available, even though the range for this system should be 30 feet max. Some of you might suggest to stick with a standard wired solution, but that is not optimal at all. Any suggestions would be greatly appreciated."

46 comments

  1. Mirrors by mfos.org · · Score: 4, Informative

    I don't know the exact setup you have, but IR with mirrors might work

  2. line of sight by zogger · · Score: 1

    --was reading here last month or so about using lasers for data transfer. Maybe find that thread/company/method, then use strategically placed and aimed mirrors to "force" a line of sight?

  3. How much bandwith? by thefatz · · Score: 3, Informative

    How much data do you plan on sending? Data over ham might work. You might be able to find some 19.2kbps wireless modems that work together. Any Ideas?

    --
    http://www.freebsd.org
    1. Re:How much bandwith? by shaitand · · Score: 2

      not if this environment is commercial, that would be illegal.

    2. Re:How much bandwith? by GoRK · · Score: 2

      Um, you know it's illegal to transmit encrypted data over amateur radio, right? Kind of shoots any security out of the water, which is probably a bad idea for his network..

      ~GoRK

    3. Re:How much bandwith? by jaeson · · Score: 1

      Data over ham might work.

      Plug that ethernet card right into a pigs ass and see what happens!!! Seriously tho, isn't this being done using packet radio?

    4. Re:How much bandwith? by Komodo · · Score: 3, Informative

      I don't think it's actually illegal to transmit encrypted data , but you have to keep a copy of the transmitted data (in the clear) for inspection by certain parties, definitely including the FCC.

      What makes it illegal is that it's abusing a part of the spectrum that's reserved and licensed for a particular type of operations (amateur radio). You need to be licensed, and the terms of the license prohibit commercial use - there's very specific rules about that.

    5. Re:How much bandwith? by Mignon · · Score: 5, Funny
      not if this environment is commercial, that would be illegal.

      He wasn't talking about ham radio, he was talking about training pot-bellied pigs to carry the data back and forth. They're very smart, you know.

    6. Re:How much bandwith? by shaitand · · Score: 2

      well damn, why didn't I think of that?

    7. Re:How much bandwith? by itzdandy · · Score: 1

      this is in the Citizen Band, theirfore not requiering FCC license. this band is also available for Commercial use as long as exsisting channel space is respected.

      if you abide by the power output and antenea hight you can use these channels for anything you want, except any encription method used is succeptable to hacks as it is LEGAL to listen and use any information broadcast on CB channels because that information is public domain while it is in transmission. if you want to use this for a network, get some really good encription.

  4. 30 feet? by Neck_of_the_Woods · · Score: 5, Funny



    30 feet that is it? Hell use sneaker net and pay a intern 5.25 and hour to run the data back and forth.

    Get a marketing intern for 10baset, a business intern for 100baseT, and a CS intern for 1000baseT. Keep in mind for the 100/1000 you going to need 2 interns for the full duplex.

    If you want to figure out what is going on with your workers just pick up one of the tri delta girls and set her to permiscous mode......

    --
    Neck_of_the_Woods
    #/usr/local/surf/glassy/overhead
    1. Re:30 feet? by grammar+nazi · · Score: 3, Funny
      Such a system is capable of high bandwidth, but the latency sucks. Especially during lunch time and afternoon break time.

      Tri-Delta girls aren't particularly good girls with regards to networks. Remember that 2 out of 3 go down (an classic joke refering to their logo). Do you really want a 66% system failure rate with your network?

      --

      Keeping /. free of grammatical errors for ~5 years.
    2. Re:30 feet? by Zerth · · Score: 2, Funny

      > Remember that 2 out of 3 go down (an classic joke
      > refering to their logo). Do you really want a 66%
      > system failure rate with your network?

      I dunno about you, but that statistic sounds like a 66% success rate to me:) Your latency is shot already, who cares if most of the packets take another 3 minutes to get there, they will be very happy packets and eager to be retransmitted!

  5. What kind of area? by duffbeer703 · · Score: 2

    Any kind of standard high-speed networking will not function in a high EMI area -- cabled or uncabled.

    I think the military uses a semi peer-to-peer wireless network between armored vehicles and some helicoptors. It is some sort of spread-spectrum technology that violates FCC rules and costs millions of dollars.

    If you cannot run shielded cable or fiber optics, you need to redesign your facility, period.

    --
    Conformity is the jailer of freedom and enemy of growth. -JFK
  6. Creative Ideas by the+eric+conspiracy · · Score: 4, Insightful

    High EMI, mmmm.

    Since line of site and wires seem to be out, maybe more creative ideas are needed. Since you don't say what the data transfer rate needs to be, I'll assume it's not high.

    For low rates perhaps sound pulses would work. This would bring meaning back to the term "ping".

    Another possibility would be paper tape. A pair of writers/readers with the tape running between them would be immune to EMI, and the tapes go around corners nicely.

    Ham radio operators have been sending TCP via radio at all frequencies in the electromagnetic spectrum for many years. While you say that the common frequencies are no good due to EMI, how about elsewhere?

    Another possibility is a hybrid system; wires where they work well, optical where EMI is high might work well.

    1. Re:Creative Ideas by n9hmg · · Score: 1

      First, I've got to say - I love the paper tape idea - reminds me of the punched metal tape used by the US military for hyper-critical information. If we're talking about kV/M, that's going to overwhelm most frontends... at least, any I could design, and most system internals, too. Paper tape could pass through a thin slot, keeping that EM out of the system, but it's extremely-low bandwidth, and uses up physical supplies. 30 meters, not line-of-sight.... How about infrared? Maybe, if the path isn't particularly high albedo, do some custom hardware work to power-up the transmitter (in this case, there's not much you can do for the receiver)... an op-amp driving a BIG HONKING IR LED is probably all you need.... In fact, do that anyway, unless it works well with stock equipment. I've heard of functional ad hoc networks using IR. If the space seen by the receivers is not brightly-lit, and not super-hot, you can probably get a nice 96kbps network that doesn't fluctuate as people walk between the nodes (enough multipath to fill).
      Incidentally... a nice feature of fiber networking is that it doesn't give a rats ass about electric fields or potentials. Can you just drop or hang fiber? That will make you all but bulletproof, with very large capacity.

  7. Other requirements: by Hubert_Shrump · · Score: 4, Funny

    It must be Open Source.

    It must look pretty.

    It can't be composed of matter.

    It can't have been thought of by a carbon-based lifeform.

    It must call me 'Capt. Studjammer' or at least 'Sir Booth Studdington'.

    Pancakes!

    24/7 always-on paradigm outside the box proactive functionality.

    --
    Keep your packets off my GNU/Girlfriend!
  8. wireless packet program by man_ls · · Score: 2

    Posted on Slashdot earlier, there is a program for Linux that will turn any radio attached to a serial port on your computer into an IP-addressable device.

    Your network's operating frequency would only be limited by the hardware in the radio, and of course the speed would be limited by the serial port, which is IIRC about 256kbps.

    Plenty fast for a "logic" connection, such as syncing two systems or for IPC between them, but transferring large quantities of data is going to get long.

  9. huh? by Anonymous Coward · · Score: 0
    (thousands of V/m at those frequencies)

    Where is this, inside a microwave oven? I wouldn't want to even work near this environment, let alone try to network in it.

    For those that care, it takes 3x10^6 V/m to breakdown the dielectric strength of air. You can get an arc at a lot less of course, depending on the metal objects present.

    1. Re:huh? by Anonymous Coward · · Score: 0

      My guess would be a naval ship of some sort. That would explain the undesirability of wires too, tough to get through steel bulkheads?

  10. Use the right frequencies by anthony_dipierro · · Score: 4, Informative

    You could use MURS along with a ham rig. Or even better if you have the time, use a licensed frequency.

  11. Why and where by WindBourne · · Score: 4, Interesting

    I am curious what environment that you need this at where you u have large amounts of voltage and signal, yet do not have control of the area? Off hand, this makes no sense unless this is needed to get around something that you are not supposed to be in. The implication is that you have a fence with high voltage and windowless wall to communicate through, like a prison. Otherwise, I suggest if you have permission/autharization, then use fiber (30 ft) or use laser.

    --
    I prefer the "u" in honour as it seems to be missing these days.
  12. GENRIP by nesthigh · · Score: 5, Informative

    Depending on the amount of data you need to transmit.. GENRIP, GENeric Radio IP, based on STRIP , may work for you. As discussed here before, it basically extends IP over any serialized radio. So, you can choose your own frequency and power.

    next

  13. power line networking by Anonymous Coward · · Score: 0

    Just a thought, but would EMI intereference affect power line networking?

  14. Modulated light by Futurepower(R) · · Score: 3, Informative

    He's right. Modulated light avoids electromagnetic noise.

    1. Re:Modulated light by Anonymous Coward · · Score: 2, Interesting

      Not if the e.m. noise is at the frequency of the light. :-)

    2. Re:Modulated light by itzdandy · · Score: 1

      but if the interference were in the frequency of IR light, it would be illegal as the FCC rules out non-IR signalling in this band of light. Unless TV Remote Controlls are being used extensively. Anyway, light is a physical thing, it is not radiation and is not succeptible to "blockout" level noise from any legal EM radiation.

    3. Re:Modulated light by unitron · · Score: 2
      "Anyway, light is a physical thing, it is not radiation..."

      Slept through Physics class, didn't you?

      --

      I see even classic Slashdot is now pretty much unusable on dial up anymore.

  15. Boy scout training comes in handy here by quintessent · · Score: 2

    I was going to say mirrors and morse code, but without line of sight, that's out. Otherwise, smoke signals should do the trick. Do you have any specific bandwidth or latency requirements?

  16. 900MHz by Koos+Baster · · Score: 1


    Eh. I'm not an expert, but isn't 900MHz the base frequency of (older) GSM phones? ' Guess telcos won't be very enthousiastic about the idea of sharing it...

    1. Re:900MHz by BenTheDewpendent · · Score: 1

      GSM 900MHz is in eurpoe and something else too. State side its GSM 1900MHz. 900Mhz is also the base to many cordless phones state side.

      I'm the generic ip over radio seems to be the best idea so far. or maybe the interns.

    2. Re:900MHz by Koos+Baster · · Score: 2, Informative

      Checked that. Europe uses 900, 1800 and 1900Mhz (a triband phone can use all of them).

      Slightly off topic; the 900 Mhz appears to use more energy, so you're probably best off with higher frequencies. This may be a reason some telcos have already decided to ditch the older 900Mhz frequency in favour of the others.

  17. Optical by Tux2000 · · Score: 1

    If you have high EMI in your environment, copper cable and radio will have much trouble, no matter at which frequency they work. You need to use either sound or light. Sound (ultrasonic) is very rare in networks (at least I do not no any solution for IP over ultrasonic), light leaves two possible solutions. IrDA (at any speed) and some mirrors, like someone else already proposed, or optical fibre. If you need to move your devices, you have to use IrDA, maybe with mirrors at the ceiling or other optical tricks (like lenses). Otherwise, just stick with plain old ethernet over fibre.

    --
    Denken hilft.
  18. *strokes cat* la-ser by awx · · Score: 3, Interesting

    Lasers modulated by the output from an ethernet port or serial port with PPP, redirected with mirrors?

    --
    Feel that power? That's mah MOUSING FINGER
  19. License by GoRK · · Score: 2

    If frequency is your only problem, then you need to figure out what frequencies are clean, then get a license to use a clean frequency and the corresponding equipment to go with it. Alvarion has a lot of stuff available for use at 3-4GHz, and there are other companies that have UHF stuff between 300-800MHz.

    Your other option is to put hook standard equipment up to modulators that will change the frequency. I have worked with devices that double the frequency of 2.4ghz devices to 5.8ghz for use in areas of 2.4 congestion for wireless isp's. There are bound to be other such devices out there.

    ~GoRK

  20. Fiber? by wowbagger · · Score: 3, Informative

    It sounds like this guy needs to communicate to some device that is at a different potential than the rest of the system.

    Have you considered using fiber? High bandwidth, pretty much immune to RF interference, high voltage isolation.

    However, if the target is moving around (some sort of industrial machine, perhaps?) this might not work.

    That's the problem with a lot of Ask Slashdots - the person asking really cannot give a lot of information about his needs, so we cannot weed out suggestions that won't work.

    1. Re:Fiber? by wik · · Score: 3, Informative

      >That's the problem with a lot of Ask Slashdots - the person asking really cannot give a lot of information about his needs, so we cannot weed out suggestions that won't work.

      Nor will s/he give information about which frequencies *are* good. If a clue isn't given, expect plenty of useless answers. In addition, there is no justification for why a wired connection isn't optimal (and how "optimal" is defined in this situation). It almost seems as if this person doesn't want an answer!

      --
      / \
      \ / ASCII ribbon campaign for peace
      x
      / \
  21. Look for vendors in bad EMF markets.. by nurb432 · · Score: 2

    Have you looked into contacting companies that make other data equipment for that sort of environment?

    Companies like Allen Bradley, Omron.. etc..

    They deal with high EMF areas all the time in manufacturing plants..

    So even if they DONT have a wireless solution, they have been asked, and prolly know of one.

    --
    ---- Booth was a patriot ----
  22. Amateur Radio by thomss · · Score: 1

    Other people mentioned this but just a little more info. Ham radio would give you more then enough choices of freq. and a good speed choice too(300bps - over 1M/s) BUT, you would need a license(not hard to get, quick 50 or so question test and $6.) Also, you wouldn't be able to use it for anything buisness related. if you just want to experminent then this would be the way to go. I've found 70cm band(around 440MHZ) to be fairly clear in noisy places and if its too noisy, you may just have to lower the speed but it still could work. Equipments spendy too though and for real high speed you might have to build some yourself. Borrow a wide band receiver and see what freqs are clean in your facility.

    -Tom

  23. Different antenna by thomss · · Score: 1

    Just thought of this too...have you tried a different antenna? a beam ant would be small at those freqs. and would cut out alot of noise off the back and sides of it. It would also direct more of your signal at the other machine. You only need to get above the noise, not cut it out completely. I believe all these devices are FM so if it can grab on to your signal it should ignore most of the noise.(like when your driving and your between two radio stations, you only get one or the other, not usually both). Just another idea that would be much cheaper. Oh, and good filtering would help too.
    -Tom

  24. Maybe... by Anonymous Coward · · Score: 1, Funny

    You should just stick to drums.

    I heard those worked really worked well in ancient times.

  25. GENRIP by gperry · · Score: 1

    Check out GENRIP from http://www.dodinc.com Provides IP connectivity over pretty much any wireless serial line device.

  26. AVIAN CARRIER PROTOCOL!!! by SirTwitchALot · · Score: 1

    sounds like a job for rfc 1149

    --
    Go away, or I will replace you with a very small shell script.
  27. blue tooth by Anonymous Coward · · Score: 0

    why not try blue tooth it dose operate in those frequencies but is prety much imune to interfierence we had a presintation at our local institute for electrical engeniers they said even operaiting around arc fenases ( very high emi)they still achived decent speedsand they too use spread spectrum as someone els sugested but at very resonable prices

  28. Consider RFC1149 by divide+overflow · · Score: 2

    It isn't ethernet, but what the heck:
    A Standard for the Transmission of IP Datagrams on Avian Carriers
    A little birdseed goes a long way....