Follow Slashdot stories on Twitter

 


Forgot your password?
Close
typodupeerror
×
Technology

Long-Range Networking 96

Posted by michael from the no-pigeons-necessary dept.
ink and several others sent in the latest Cringely column where he discusses creating a long-range 802.11b network using high-tech tools like a telescope and bribery. Sean Clifford sent in guide to creating your own long-range antenna: "Have some old PrimeStar hardware laying around? Do something useful with it by turning it into an IEEE 802.11 wireless networking antenna. This electrical engineering professor uses an Airport, but any access point should work fine." If you're at all interested in this sort of stuff, get involved with one of the community wireless networks springing up.
This discussion has been archived. No new comments can be posted.

Long-Range Networking More

Long-Range Networking

Comments Filter:

  • Re:Thought the telescopes..... (Score:1)

    by Anonymous Coward
    A number of amateur radio types have migrated to experimenting with optical communication using lasers and other lighting sources. Two approaches seem particularly pratical low pressure sodium lamps and low pressure mercury vapor. The sodium is neat because almost all of its output is in two closely-spaced emission lines, making filtering easy and cheap. Also, almost all commericial sodium outdoor lighting uses high-pressure bulbs which have a different, non-interferring, output spectrum. The mercury/UV approach is neat because most of its output is the "solar blind spot" created be the ozone layer. This means its quite dark at these freqencies even in broad daylight, even if the sun is in the field of view of the optical system. steve schear@lvcm.com

  • Re:FCC Part 15 rules (Score:1)

    by Anonymous Coward
    Sorry, but it is legal

    You only need FCC Part 15 certification if you will be selling or manufacturing the devices.
    The FCC allows individuals to make up to 5 devices without applying for Part 15 certification.
    This is all covered under the Part 15 section for homemade electronic devices.

  • Re:distance/legal (Score:1)

    by Anonymous Coward
    >His set-up is most certainly illegal.

    Wrong. Legal EIRP is +36 dBm on multi-point links. The power limit is +30 dBm (1 Watt) at the product's connector. You can use higer EIRP on point-to-point links. See Part 15.247 for more info.

  • Wireless Laser (Score:1)

    by Anonymous Coward
    I once built a similar thing with a laser, you can have a look at it: http://strike.wu-wien.ac.at/~dusty/projekte/laserl ink We managed to set up a 128kbit connection over ~2km, longer distances would have been possible. Moreover it was cheap: Around 250$ for the whole thing. Our next goal are 10Mbit but we suffer from lack of time. The best thing about this is that there are (at least in Europe) no problems with FCC regulations at all...
  • This message brought to you by TechsPlanet.

    Just got done reading a story about how you can modify equipment laying around to do what you want, and have the urge to do it yourself? Well, come on down to TechsPlanet.com, where we will sell you the same stuff, done "right", for signifigantly more than it would cost for you to do it yourself. And for a limited time, we'll even throw in a free Slashdot subscription with every purchase! Be the first troll on your block with a wireless LAN and a Slashdot UID over 400,000!
  • I'm guessing you make a comission, and you have access to the web server's referrer logs?

    You forgot to have a referal link to something as a sig.
  • Figures.

    -adds *techsplanet* to spam blocking list, as that seems to be next-
  • I doubt this. Cite who, when, where, etc.

    What the fuck is this "must wait 20 seconds" shit. what a load of crap.
  • You don't know what you are talking about. The FCC allows up to 1 watt of effective radiated power from the antenna for us to use in the 2.4Ghz band...the wavelan (aka orinoco) transmit only 0.03 watts.

    Goto the FCC's link: http://www.fcc.gov/Bureaus/Engineering_Technology/ Notices/1999/fcc99149.txt [fcc.gov]

    Snip:
    4. Frequency Hopping Systems. Section 15.247 of the rules permits frequency hopping
    spread spectrum systems to operate in the 2.4 GHz band with a maximum output power of 30
    dBm (1 watt). The rules specify that frequency hopping systems operating in this spectrum must
    use a minimum of 75 hopping channels with each channel having a 20 dB bandwidth not
    exceeding 1 MHz. The average time of occupancy on any frequency must not be greater than 0.4
    second within a 30 second period.

    /snip

    BTW, where do you get your $150,000 nonsense? Since when does the FCC have time to bug private citizens unless it is intefering with airport/medical/government communications??? This is why the ham radio freq's have become mostly self-policed. Can you please provide a link to an authortivative source to back up your claim?

    Just my $0.02 worth of flame. Sry if I have come on a bit strong. It has just been a long day....don't take it personally.


    ---
  • amen to that! the only diff I see is that these people playing around w/IEEE802.11b most likely do not have their ham license.

    AND, even if they did, they wouldn't be able to use the 2.4GHz band...they would have to use the band that is a bit lower...something like 2.1Ghz if I remember. IE: they are restricted to the ham bands.

    BUT, otherwise, I couldn't agree more.

    ---
  • Big deal.

    I have Road Runner cable modem service through Time-Warner Cable in Milwaukee. Officially, they don't allow home networks, NAT boxes, etc. Unofficially, they don't care as long as you don't abuse the service. The guy who installed the service smiled and checked the "No" box next to Home LAN as we plugged the modem into our hub. We've run five computers off our one connection for nearly a year now and nobody cares. Odds are the situation is similar in this case.
  • I'm the Net Admin for a WISP in the North East Georgia Mountains, and Cringly spent WAY too much for his gear / connection. He could have used some $400 Teletronics APs and two $80 25db Grid antennas to accomplish the same thing for just about $560.

  • Whoops, total is $960..

    bleh. I failed math.
  • Yeah, you'd do a lot better with lasers, and it certainly is not the first time someone thought of it. The problem with lasers is (a) rain, and (b) thermals in the atmosphere, which make the beam wander around. This probably happens to RF, too, but since the beamwidth with a laser is so small, it really messes things up. Just ask youself why stars twinkle...
  • Yes, but..

    One thing Cringely did not go into was the service agreement he or his co-conspirator signed when they got PacBell's DSP. Almost certainly it is verboten to resell connectivity if you are on a consumer DSP contract. Now that he has blurted this all over the 'net, I expect to see a future column on how the law and PacBell comes down on him.

  • What the hell fun is that? Granted it sounds incredibly functional, but not a single warranty is voided in the process, nor does it involve the use of any common household items.

  • I think you're right, and the Cringeley article is therefore, unfortunately, crap.

    We looked at doing the same thing, and found that there is no way to make two AirPort base stations talk to each other. In the end, we put a Linux router on the other side of the link. We were using an earlier firmware on the AirPort (I think 1.2), but the release notes for 1.3 don't mention any new capabilities along these lines

  • Early on when they were setting up networking at Xerox's PARC, they used four telescopes and a pair of lasers and detectors to network between two bulidings separated by I-280. Apparently several people reported UFO or strange object sightings when the red laser beams would sporadically show up in the fog!

    I think this story is mentioned in Michael Hiltzik's Dealers of Lightning [fatbrain.com] (fatbrain link) history of PARC, which is a worthwhile read.

    Regards, Robert

  • I noticed a distinct lack of any math on the primestar link page - is making an efficient waveguide really as easy as stuffing the center conductor of some coax a ways (1/4 wavelength, right?) into the side of a tin can and saying "that looks about right?"
  • You probably want to set up a digipeater. This might be as simple as putting a solar-powered access point on the mountain and aiming cards at both sides at it.

    If you want to take the purely passive approach, find a way to keep a large area of 1/2" or finer chicken wire in a very flat plane and use that for a mirror. You might want to temporarily attach an optical mirror to it and use a laser to check the angles. Make sure this setup won't sway in the wind.

  • Yes indeed, perhaps a relay contact got fused. If you're handy with a soldering iron you might be able to fix it.

    Be thankful the modem failed in an obvious way. It's much better than having it fail just enough to be a nuisance, and spending an hour on the phone with a technician.

  • Isn't line of sight basically a 2 node network? Does it really make sense to use a protocol that worries about polling and collision detection when RTR and RTS should handle it just fine?

    Does your post have any HTML in it? Did you specify "Plain Old Text", or "HTML Formatted"?

  • Or maybe support one of the many free wireless communities.

    Personal Telco's List of Wireless Communtities [personaltelco.net]

  • I'd just like 802.11b to work 30 stinking feet in my house. It has to go through one lousy wall and it's impenetrable (no, no metal in there). Even in the same room with no intervening objects, a 20-foot separation cuts the signal strength in half, according to the software 'meter'. I must have a bitch of an RFI problem or something. Either that or Linksys products are crap (my router works great, though).
  • Do they have internal or external antennae? Do you have any other 2.4Ghz products? Cordless phone? Video transmitter? Does your microwave leak?

    The WAP has dual external 'diversity' antennas. The PC Card (Linksys WPC11. Not sure of its output power.) has an internal antenna. Orientation is a problem. I can occasionally get a weak signal if I orient the card vertically (by holding the laptop up), so I may try a card with a vertical external antenna instead. I tried putting the WAP antennae horizontal to attempt to match the polarization of the PC Card, but no luck. I've got several 2.4 GHz products, but I've tried shutting them down, to no apparent effect. The microwave isn't in use.

    what's the wall made of?

    Standard stud-and-drywall. My 2.4 GHz video link goes thru it like it isn't there (admittedly, it has a directional antenna and probably more power).

  • Time to get out your Rhode & Schwartz spectrum analyzer and check out what's goin down in your airspace ;). Maybe a neighbor has his experimental microwave weapon pointed at you.

    I'm afraid that you may be right. It wouldn't surprise me to find that somebody has a raft of 2.4 GHz stuff running next door. Might need to deploy my anti-radiation missiles. What I was actually thinking of doing was to cobble together a crude yagi out of wire, just to see if directionality might help. The location of the WAP is in one corner of the house, so if I could get a fat lobe going diagonally, it might help. If that worked, then I could consider a commercial antenna.

    External 'diversity' antennas? Is that when there are two omni's, one on either side of the unit? Gah.

    Yep, that's the setup.

  • Our US government supports various monopolies. There are laws which protect the phone companies, power companies, and cable companies.

    If my solar panels can produce more power than I can use, federal laws prevent me from selling it directly to my neighbor, unless my neighbor is the government sanction local power monopoly. Feds will come with guns if necessary to shut me down.

    There are many regulations that the Gov't has around telecommunications. They have been put in place over the last 100 years to keep the monopolies propped up. It wouldn't surprise me for their to be a "providing telecommunications service without a permit." I'd be surprised to for it to be a city law though.

    I would like to see that posting get validated too. People don't realize how much gov't control their is. When the news talks about "deregulation" it's really a big joke. They merely change the regulation, or perhaps reduce the level of control. They never actually get rid of the regulations.

  • They really need to update that one with bio engineered avian carriers or something sheesh ...
  • To go much farther you either need more power (which may tick off our friends at the FCC) or directional antennas, like Cringely used, with clear line-of-sight.

    Any ham will tell you that it's not the power... it's the proper matching of the antenna.

    I have a friend here in Singapore who uses 10 watts to talk to his buddies in East Coast USA!

  • While it may be "l33t" to hack old primestar gear to make a shot, you're not coming up with anything new..

    OK, it may not be NEW, but it is cool. And it is new to most readers.

    Some radio standards have concrete distance limits. For instance, the GSM cellular standard works on time division multiplexing. If your station is over 22 miles away, it simply won't work. Your "packets" will arrive at the base station too late, and they will clobber the guy in the next time slot (speed of light, and all that).

    My point is that it's sometimes cool to understand a little bit about 802.11 RF standard, so we know its limitations and "fringe uses".

  • I'm saying that he is getting the 20db gain (if any at all) from the dish, not from the feedhorn.
  • OSDN wants at least 10k$ to advertise with them.

    Look for our Computer Shopper adds starting this September.

    And yes, we have done VERY long distance hauls, 40+ miles. it really depends on the tower height. We have even set a 160 mile haul using space on 900' towers....
  • This little commission called ownership... :)


    Donald Beckman
    Techsplanet.com/wlan
  • Seriously, many people will read this and not want to do it, rolling your own will break the law in a way you CAN BE TRACED!!!! Unlike the normal slashdot stereotype, many of us would rather buy a legal system....
  • but, our systems can get so powerful that you can melt things with your microwave transceiver! Now that's fun!!!
  • with 100mw, you can use a 19dbi or 21 dbi antenna on the TX side. Using a larger antenna also gives you more receive area, so you can pick up fainter signals from further away.


    Check out http://www.techsplanet.com/wlan/calc.htm for a link calculator. a ? Will appear next to the "Peak transmitted EIRP " of a non-legal setup.
  • Our service has to pay TIF taxes (which we charge our users for), but as long as we do not provide phone service, voice over IP, then we do not have to register as a CLEC

    Donald Beckman
    Pelican Bay Internetworking
    www.techsplanet.com
    wLAN Systems:
    www.techsplanet.com/wlan
  • Using discovered hosts, this could be a sweet way to share files with people.

    I submit that this isn't used for real networking, but could be used to join two home LANs across the neighborhood.

    By using only `discovered` hosts - and connecting to only other neighborhood hosts you could set up a great file network. Imagine, none of your data would actually need to be sent over the wires, just beam it back and forth.

    But I guess as soon as other hosts `discover` your rouge LAN it would be hammered.

    I've always wanted to share my cable connection with my [little] sister - she just lives 3 miles away.
  • What if there is a mountain in between. Is it possible to construct something that for instance reflects in the atmosphere? My neighborough town is probably getting DSL, it's only 5 km away, but there is a mountain inbetween. The first long-long-long distance transmissions were done that way.
  • How about removing the telescope and the lightbulb and just having a laser and some sort of sensor. Cheaper, more reliable and so on. On the other hand, what if it's a hazy/foggy day?
  • I am quite certain that the power limit is 1500W PEP. But I dont think anyone would commplain that that is not enough. After all we are not trying to reach the moon. (EME networking might be fun though) The real prob is humidity -- it absorbs the microwave RF.

    Remember, when you are downloading MP3's, you are downloading communism!!!
  • Wow, retro...

    The juice can antennae at the bottom of the primstar link [wwc.edu] is just the latest evolution of the old tin can telephone [dsokids.com] that I remember from my youth.

    --

  • How else do you think NASA controls it's sattelites remotely?

  • And HOW is this news?! (Score:2)

    by Anonymous Coward

    There's a few wireless ISPs that have been using 802.11 for long distance shots for quite a long time..

    Longest one I ever did was 13 miles with no amplification, using standard 24 dBi antennas. That was over 2 years ago (yes I know it wasn't 802.11 back then, but basically the same 2.4gHz wireless networking gear.)

    While it may be "l33t" to hack old primestar gear to make a shot, you're not coming up with anything new.. for a couple hundred bucks you could buy two 24 dBi antennas and easily make line of sight shots over 10 miles with no amplification.

  • Re:distance/legal (Score:2)

    by Anonymous Coward
    At close distances, it'll punch through quite a bit. I run an Airport card in a G4 as my "base", and a stock Wavelan Gold card in my laptop as the remote system. Both run Slackware, and are configured in the "old ad-hoc mode" which lets you just ifconfig the interfaces with reasonable numbers and it just works.

    As for distance, I can go about 150 feet in most directions, including getting out of my house, crossing the yard, and getting through the walls into another house. It starts getting touchy inside the next building, but that's life. The point was to have total coverage inside my place and the yard immediately surrounding it, and I have succeeded.

    The next step is to get some antenna action happening so I can light up even more space with the signal. Right now, everything is stock and it's still amazing just how well it works.

  • Even if the airport is just a lucent card in disguise. The antenna and base have not been certified together. A no no.

    Not true. the law is a bit complex, but it isn't definatly illegal to work to modified one off equipment. It is illegal to sell such equipment, but the requirement is that modified equipment be professionally installed. If your not breaking any noise or power requlations and your installation looks nice, then you have a reasonably arguement that your setup was professionally installed by you.

    If that will hold up in court depends on the judge and jury, but generally it won't go that far because the FCC has better things to do then worry about someone on a license free band who isn't breaking power limits.

  • You said:
    Also is it legal? I could just see the FCC knocking on someone's door for this.
    Also what about diameter, do you have to be in direct sight of it, or will it work though walls and if you are off to the left 50 feet?

    The article goes into both of these:
    1) the fcc only cares about signels over a certain strength. This lives comfortably below that threshold.
    2) Yes, this must be line of site. Cringely went looking through three different telescopes to find someone to work with.
  • Why would you think there's any HTML in my post? Does it show up that way in your browser or something?
  • Could you name the community where this happened? Or better yet provide a link to a news story about it? I'd like to know more, if only to avoid ever passing through there.
  • If you really were the time would be 25 or 6 to 4.
  • Isn't line of sight basically a 2 node network? Does it really make sense to use a protocol that worries about polling and collision detection when RTR and RTS should handle it just fine?
  • Cringely's satellite setup had a crap uplink which I assume was dialup. Assuming the satellite downlink is 2 Mbps, and the modem 33.6 Kbps on the uplink, the ratio between downlink and uplink is about 60.

    On a big download, he would receive Ethernet style 1500 byte packets on the downlink, and send 40 byte TCP ACKs on the uplink. Unfortunately 1500 / 40 is 37.5 - he would need to send 2,000,000 / 37.5 = 53 Kbps of ACK packets over his 33.6 modem, so his downlink is reduced to about 1.3 Mbps when he's using TCP.

    This is a bit of an artificial example but it does illustrate why extreme asymmetry is a bad idea with TCP, and of course any other traffic on the uplink would slow things even more. There is some work on optimising TCP to solve this, e.g. ACK filtering and ACK reconstruction.

    BTW, high latency still sucks because TCP's ability to ramp up its congestion window after some packet losses is controlled by the round trip timer - high latency means a slow ramp up after every significant loss, which is particularly bad if you go into slow start and have to start from one packet again.

    In this case it was probably more latency than asymmetry because his link was unlikely to be 2 Mbps.
  • I've set up similar networks. However, I managed without the dish, and instead of coax I used string.

    Strangely, it was much easier to use it for VoIP than Email, Web, or Quake.
  • The power limits are further constrained by the requirement to use the minimum power necessary to achieve communications, and the need to control uwave RF exposure to people near the antenna.

    Plus generating that much power at 2.4 GHz is EXPENSIVE as well as unecessary.
  • True, but the FCC regulations for the Amateur Service place limitations on the content you can send, use of encryption, and other restrictions that may make "souping up" an 802.11b system more useful.
  • What kind of losses do I get firing through trees?
    as many cant put up 100 foot towers to get over the trees here in suburbia If my target is less than 5280 feet away will 22DBi gain burn through the trees inbetween?

    (I'd calculate it but I cant find my ARRL antenna handbook.)

  • Agreed. In support of this, we provide a free courtesy LANRoamer account to anyone who runs an open access point that meets the other LANRoamer criteria (up most of the time, in our coverage maps, technical contact info provided, etc.).

    I don't think this policy is on our web page yet, but it will be soon and we publicly stated it in our presentation at the last sbay.org [sbay.org] meeting, and possibly in our presentation at the last Bay Area Wireless User Group [bawug.org] meeting.

  • Goofy. Aside from a bad antennae connection, perhaps you should try replacing one/both units?
    It only takes a small nick in the antennae connection to screw things up horribly, though... 2.4ghz is quite sensitive to that.

    Time to get out your Rhode & Schwartz spectrum analyzer and check out what's goin down in your airspace ;). Maybe a neighbor has his experimental microwave weapon pointed at you.

    External 'diversity' antennas? Is that when there are two omni's, one on either side of the unit? Gah.

  • Cool.

    Good luck with the yagi. I know we've used them lots and they make a huge difference.

    ANy place you can borrow a spectrum analyzer? One of the local hammies or a cellular tech will have one in his truck.... maybe for a case of beer a cellular guy will come over and sweep the place?
  • If you are in some country with a monopolistic government, some small carribean island, then I might believe this.
    As for the US, I doubt it.. I don't believe there are civic laws regarding providing telcommunications.

  • You have an RF problem somwhere.A 20 foot unblocked airspace should provide no discernable difference, at least, not on the software meter they show.

    Linksys tends to be reliable.

    Do they have internal or external antennae? Do you have any other 2.4Ghz products? Cordless phone? Video transmitter?
    Does your microwave leak?

    Also, what exact model of card are you using? What's it's output power (not that it'll be the problem)

    Also, what's the wall made of? 2.4Ghz doesn't like going through concrete either. It doesn't like going through much of anything, actually. Thin wooden walls in the house will be okay, thin partitions in offices are okay.... thick wooden walls present a large barrier.
  • - 802.11 has specifications for doing exactly this kind of link
    - Yes, a customized protocol tuned for exactly the distance you are going would be more efficient, but you'd have to write it, and you wouldn't get that much of an improvement.
    - 802.11b does not do collision detection, it does collision avoidance (has to do with timers). You cannot detect a collision reliably in RF, especially at these distances.
    - The lack of collision detection is one reason why you need rts/cts and other types of handshaking. The other is radio noise.

    - 802.11b doesn't poll (I don't think.)
    - Adjustments to the inter-frame gap in 802.11b (if memory serves) to account for increased distance can cut down largely on the # of collisions, but decreases the available slot time, so a transmitter may have to wait longer to transmit, but will transmit faster once it goes.

    - This can be extended into 3 node, or other configurations where not all stations can see each other; 802.11b can deal with this as well.
  • If he is right about the antenna being 21dB then it is most likely to be illegal.

    Most 802.11b cards put out about +12 to +15dBm, some more than that. +12 + 21 = an effective radiated power of +33dBm. The FCC limits for the 2.4GHz band where 802.11b lives is 1W maximum, i.e. +30dBm.

    His set-up is most certainly illegal. The FCC cares about the effective power, not the power fed to the antenna.

    Also, if you are using a +21dB antenna and you are 50ft off, it is probably not going work. A +21dB antenna is going to have a _very_ narrow beamwidth.

  • Note that the FCC limits the radiated power. If your antenna has a gain of 21 dB, you MUST limit the maximum power of the radio to 5 mW. (4 W EIRP maximum limit). The limits abroad are typically even stricter.

    So at some point it makes no sense in making more directional antennas, as you have to decrease the power level accordingly.

    See this link [cisco.com] for more information.

  • When I first read this article.. I was chuckling at his ingenuity..and determination...sweet... Then I thought..To help subsizde his monthly cost and his inital outlay..talk to his neighbors.. I'm sure the signal isn;t so omnidirectional that the radio signal goes straight to his house without any spill over.. He can set his neighbors up on 802.11b as well,, charge them a modest fee and he can be come is own neighborhood ISP..
  • I'll be damned, I've never heard of that. That's insanely cool!
  • True, but I can't believe they even let you try, without at least a ham license. Very un-governmental of them. :)
  • This is all covered under the Part 15 section for homemade electronic devices.

    I've never heard of that. What subpart is it under? That's cool, if true.
  • I assume he didn't spend $2K just to play Doom.

    Oh, I don't know. I spent way more than that building this [qsl.net] just to play Quake...
  • Plus generating that much power at 2.4 GHz is EXPENSIVE as well as unecessary.
    Nah, dirt cheap. It's called a magentron, and you can get one for $100 at Wal Mart. ;-)
  • is it a good idea to intentionally "de-tune" an antenna, to reduce it's output to just what you need (as opposed to using an attenuator).

    I would say no. There's only a couple places that power can go:

    • Back down the coax to the card (higher standing wave ratio). This may or may not matter for 30 milliwatt 802.11 transcievers but it does matter for anything over a couple watts.
    • A direction you didn't want. Less directional antennas don't have as much gain in any one direction.
    • A harmonic you didn't want. Luckily there's not much to interfere with at the higher frequencies yet.
  • 802.11b really isn't designed for that long a path. 20 miles is 0.1ms of delay, which is long for that protocol. It will still work, but you won't get anywhere near full bandwidth. Too much time is wasted contending for airtime.

    Point to point microwave links [slashdot.org] for such things are available, although they're not common consumer items.

    Incidentally, links like this need some extra gain margin if you want them to keep working during heavy rain. Rain passes gigahertz frequencies about as well as it passes light.

  • ...making your private network availible to even more people.

    First lets work out the security holes (of wireless networks) and make sure the admins can properly configure them before proclaiming wireless networks "the coolest thing since sliced bread".

    There are way too many companies out there who have just lept into the wireless market with both feet and later found out that any Tom, Dick and Harry had access to their network (example: *cough* Sun *cough*)

    - Can someone help me off of my soapbox?

  • What is the distance on these things (prime star hack)?

    Also is it legal? I could just see the FCC knocking on someone's door for this.

    Also what about diameter, do you have to be in direct sight of it, or will it work though walls and if you are off to the left 50 feet?


  • Section 15.23. "Equipment authorization is not required for devices that are not marketed, are not constructed from a kit, and are built in quantities of five or less for personal use."
  • www.nycwireless.net [nycwireless.net] is a free public 802.11 project in New York City. We already have two nodes active with more coming soon.

    A complete list of free wireless projects at PersonalTelco [personaltelco.net].

  • This project sort of made sense to me until I read this:
    But mainly it gets me two things I really value -- reduced latency and a useable Internet time signal. I'll never have to set my computer clock again.
    I wonder if RXC knows that you can get a perfectly good time signal from a $50 GPS?

    And will someone explain to me why latency is important? I assume he didn't spend $2K just to play Doom. For a server? If I ever spring for a dedicated server, it will live in a web farm, not behind an unrealiable 1 megabit DSL link.

    __

  • Actually, the last time I made an inquiry to my local wireless internet company about their coverage in my area, I got into a conversation with the person on the other end of the phone about their network topology and so forth. It turns out that even though their hardware is only rated to shoot the signal up to 20 miles, they've actually gotten it to go farther than 25.

    Oh, and don't you think you should just buy banner space on Slashdot instead? It's probably a lot more effective.

    --
    < )
    ( \
    X

  • If you'd like to see just how much an ISP cares about policing its network, go to grc.com [grc.com] and read his account of tracking down Denial of Service attacks on his system. He had clear evidence of hacked machines running as zombies, hijacked dial-in accounts, etc. and got exactly zero interest from the ISPs on whose customers were being exploited. The entire story is a fascinating read on how he tracked down the perpetrators of these sort of attacks and got it stopped (no thanks to the ISPs or law enforecment).
  • Can you really use two Airports in this way? Cringely writes that he makes his connections using two Airports. Here is a quote from the article:
    The set-up is simple. I had to buy a new Apple Airport hub for each end of the link.
    The 802.11b access points I have played with don't seem to directly support this. Most vendors make you buy a different product, even though the hardware is probably exactly the same, and the only difference is firmware on the radio card and box..

    I checked a recent link [apple.com] about designing Airport nets and some other Airport sites [akamai.net]. They don't reveal any clues about this being possible.

    I checked the Cisco/Aironet site. For the 340 Series WLAN gear they have a product called an Ethernet Bridge [cisco.com] which is what you would use to do what Cringely said he did. This is different from their Access Points [cisco.com] and Base stations [cisco.com] . A Cisco base station looks to be equivalent to an Airport.

    Does anyone have any references on how to do this with an Airport? I'm finding Cringely's story a little hard to believe.

    Kevin

  • Why not use a standard LNB module instead of a juice can? Is it not possible to transmit with one of those? How about one from those 2-way dishes (is there even any difference)?

    Microshaft still OWNZ JOO! [msnbc.com]
  • [blah, blah, blah from college EE text...] I'll bet that his supposed 20db gain is due more to the colmination effect from the dish itself (it is acting like a focusing lens) than due to proper waveguide design.

    If the guy gets a 20db gain from his setup, then who cares? He got it to work, he's not selling the equipment to you, if you try it YMMV, etc, etc.

    So, either you say "this guy isn't getting a 20db gain" and prove it, or drop the whole impromptu /. EE course and admit it just plain works! Obviously, he's not too far off the calculations (too far being defined as not getting his needed 20db gain from the setup) or he's lying about the gain. Which is it?

  • When done correctly, you can get a 20 mile + link setup. The operative word being CORRECTLY!!!
    The tin can antennas are not correct, nor is altering a wLAN card to do something it isn't ment to do. Use a Cisco AIR-PCI352 or AIR-LMC352 100mw card for your long range wireless links, and the correct antenna + cable.

    We do this legally and professionally; we have many antennas, cables, amps, and other accessories. We also have complete long-range pre-built systems

    Donald Beckman
    Techsplanet.com
    Wireless sales @ www.techsplanet.com/wlan
  • Get a Technician Ham License and you can use lot's of power and be legal, 802.11b is in a ham band(2.4 Ghz I believe) please check this site as it has lot's of information for hams and people that aren't hams too concerning highspeed wireless data.http://www.qsl.net/kb9mwr/projects/wireless/ [qsl.net] John KG4AKV
  • I still think RFC1149 [ietf.org] is a better, and cheaper alternative! ;-)
  • I just made a "double quad" antenna for my airport, but I haven't had a chance to test it yet. It was pretty easy to construct, but I wasn't sure *how* critical the measurements were - I just used some RCA couplers I had lying around to extend from the reflector plate - they're not the right length, but I wonder how much it matters if you're off by 10% - or say 25%?

    I'm only trying to go 200 yards line of sight, so I just want a very simple/cheap antenna - I figure if even the simpler 802.11 designs are made to go 1Km or more, then I can afford to botch it a little.

    Also I was wondering - is it a good idea to intentionally "de-tune" an antenna, to reduce it's output to just what you need (as opposed to using an attenuator). I don't want the FCC knocking on my door because I'm interfering with all the cordless phones in my neighborhood.

    Here's a link to some 2.4GHz antennas [dabsol.co.uk]. Has anybody here experimented with these designs?

    ---

  • Re:FCC Part 15 rules (Score:3)

    by Anonymous Coward on Saturday June 30, 2001 @04:16PM (#117682)
    While the URL you site is an *excellent* reference and breakdown of the regulation, it overlooks an important fact that almost all the wireless efforts seem to gloss over (including guerilla.net, NY wireless, Seattle wireless):

    It is illegal to *run* a setup using an antenna (omnidirectional, yagi, dish, parabolic mesh), cable (you soldered it up), OR amp, if any part of the setup is something you slapped together.

    The setup you purchase, from the card to the cable (to the amp) to the antenna must be certified altogether. iow, it must be sold specificly and unmodified, or must have been certified for installation by an FCC licensed installer.

    If you have a WaveLAN/Orinoco card, you have to buy Lucent's antennas or antennas approved to work with Lucent's card. If you construct your own antenna, you can use it on your card as long as it's been approved.

    You see a lot of sites about soldering some coax cable, constructing antennas out of wire clothes hangers, or using adapters to fit the external antenna plugs on some of these 802.11b cards...they are all nifty, cool, etc., but if they run their setup and did not get it checked by an FCC licensed/approved installer, it's an illegal setup.

    I'm not saying that it won't be within regs. I'm not saying these setups are bad and evil. I'm saying that, if someone finds out, whether by social interaction (you talk) or stray emissions (neighbor doesn't like your ugly dish), and the FCC comes knocking, you can be fined up to $150,000. Not cool.

    When people hear this, they get all huffy and puffy about "how their installation" works and isn't harming anyone, or how they won't get caught. Blah blah. I'm not commenting on whether it works or not. Don't flame the messenger--that's what the regs state.

    btw, this issue has been brought up by others in other /. posts after stories involving 802.11b setups, esp. the earlier ones. Look for some of the +5 posts.

  • Re:RF Engineering??? (Score:3)

    by dattaway ( 3088 ) on Saturday June 30, 2001 @02:00PM (#117683) Homepage Journal
    this reminds me of those folk who THINK that putting a bunch of aluminum around the ends of their (old) TV antenna's actually make the thing work better.

    Hell, if it works, fuck the math and milling out expensive blocks of aluminum! I used to strategicaly put an empty soda can next to my wireless card when too many cars were parked between my house and my neighbor down the street. When it started to rain, we had to break out the aluminum foil.

  • Power limits and the FCC (Score:3)

    by pongo000 ( 97357 ) on Saturday June 30, 2001 @03:46PM (#117684)
    Technically, there are no power limits for certain amateur radio license holders in the 2.4GHz range other than those that address interference to the primary allocation holder (amateur radio is secondary in the 2.4GHz range). Plus, amateurs are permitted under FCC regulation to modify their equipment (including commmercial equipment) as long as the mods do not generate interference above certain limits.

  • RF Engineering??? (Score:3)

    by rongage ( 237813 ) on Saturday June 30, 2001 @01:54PM (#117685)
    This is quite interesting...

    SWR - who needs to be concerned with that...

    Wavelength of the feedhorn - who needs to be concerned with that...

    There is a reason that standard feedhorns are patterned out of machined steel/aluminum - dimensional control. The size AND shape of the feedhorn is critical for proper (and efficient) transfer of energy from the wire to the waveguide/air. Even the shape of the exciter element (the 1.15" wire soldered to the end of the RF connector) is quite important.

    Then again, this reminds me of those folk who THINK that putting a bunch of aluminum around the ends of their (old) TV antenna's actually make the thing work better. It works for one articulate idiot so the rest of the world jumps right in. Can anyone say "Lemmings"?

  • Shameless commercial plug for LANRoamer (Score:4)

    by Adam J. Richter ( 17693 ) on Saturday June 30, 2001 @02:05PM (#117686)

    Apropos to community networks, I'd like to make a shameless plug for a GPL-compatible open source gateway [lanroamer.net] and settlement system, downloadable from ftp://ftp.lanroamer.net/pub/lanroamer [lanroamer.net]. The backend software will also be released under GPL later this weekend. There is obviously a business behind it, but the software and the ideas are relevant to this article in their own right. The basic idea is that whoever puts up an access point gets free use of the everyone else's access points and a significant share of share of the revenues from paying customers (expected cost to paying customers: $20-$25/month). We are in the process of setting up a sourceforge area [sourceforge.net] for the software as well. Finally, if you're curious about future development direction, you might want to check out the current wish list [lanroamer.net], although completely different additions to this list and, better yet, contributed code are welcome.

  • Thought the telescopes..... (Score:4)

    by Restil ( 31903 ) on Saturday June 30, 2001 @02:07PM (#117687) Homepage
    At first when I read the summary of the article, I was thinking that he was going to somehow use those telescopes to actually DO the networking instead of just using them to find a house.

    In fact, I kept thinking this beyond the point where he was talking about street lights and stop lights.

    I wonder, how feasible could such a system be? If I had a high powered light bulb (or some other convienent light source) that could switch on and off very rapidly, then at the other end of the "network" have a telescope with a CCD camera watching the light, as long as the telescope never moved, data could be sent at a rate equal to the speed at which the light could be toggled.

    I can see a great number of potential obstacles to this which would make it difficult to use in a production environment. But it would be a cool hack nevertheless. :)

    -Restil

  • Low Cost Wireless Network How-to (Score:5)

    by Anonymous Coward on Saturday June 30, 2001 @02:19PM (#117688)
    Here is a Low Cost Wireless Network How-To [gbonline.com].

    It covers the technical details that the other wireless networking sites miss. It has amplifier ideas and schematics and external antenna mods using old MMDS dishes. It even has CGIs [gbonline.com] for preforming wireless link analysis.

  • Airport is great for these hacks (Score:5)

    by jht ( 5006 ) on Saturday June 30, 2001 @02:01PM (#117689) Homepage Journal
    Airport is ideal for this sort of project - the base stations are relatively inexpensive, and they can be easily configured without a Mac - one fellow wrote a Java-based configurator app. They also work with pretty much any wireless card out there, AFAIK. I also use a Linksys with mine for one of my Wintel boxes.

    I took my Airport and added the Lucent range extender antenna (about $60), and simply dremeled my base station to accept the antenna mount. I've been using it for about a year and a half now, and it gives me an effective range of about a quarter mile (it helps that I mounted the whole rig on an outer wall upstairs in my house) when used with my iBook.

    To go much farther you either need more power (which may tick off our friends at the FCC) or directional antennas, like Cringely used, with clear line-of-sight. You're subject to all limitations of the 2.4 GHz band, though, and a lot of current cordless phones run in that range - it can mess up 802.11 signals somewhat.

    I stick to 900 MHz digital phones partly for that reason.

    - -Josh Turiel

  • Re:RF Engineering??? (Score:5)

    by Lumpy ( 12016 ) on Saturday June 30, 2001 @06:41PM (#117690) Homepage
    Oh man Ham radio operators have been using soup cans and other cans for feedhorns before commercial "experts" came in.

    yes ther eis a reason for having everything over-engineered as you stated above.. It's for efficency.. This guy couldnt care less as long as he got 20Db gain... a real ham with a better source of soup cans and a 5 foot surplus spun aluminum dish could get 50-60db gain out of a properly sized soupcan.

    Go buy a ARRL handbook, and the ARRL antenna handbook. learn about Dish antennas and how to do it in your garage as good as a multi-jillion dollar company with computer machined parts.

  • FCC Part 15 rules (Score:5)

    by Adam J. Richter ( 17693 ) on Saturday June 30, 2001 @03:39PM (#117691)

    Tim Pozar's [mailto] page The FCC's Part15 Rules and Regulation and 802.11b emissions in the ISM 2.4GHz Band [lns.com] discusses this and has links to the regulations and other useful references. Look for the section titled "Fixed, point-to-point paths and get even more power."

Slashdot Top Deals

If all the world's economists were laid end to end, we wouldn't reach a conclusion. -- William Baumol

Close