Follow Slashdot blog updates by subscribing to our blog RSS feed

 


Forgot your password?
Close
typodupeerror
×
Hardware

Cheap Wireless 802.11b Bridging 122

Posted by Hemos from the making-them-work-together dept.
eggboard writes "You can bridge two wired networks using two cheap Linksys 802.11b access points. This isn't exactly new, but the article I wrote, which just went up on Friday, describes in excruciating detail how to configure the units. The big news is really price: the WAP11 described is about $185 with a manfacturer's rebate. Using higher-gain antennas than the ones shipped with the WAP11 and/or tuning line-of-sight access, you've either figured out how to hook up a neighborhood of separate wired networks, or how to link multiple offices cheaply."
This discussion has been archived. No new comments can be posted.

Cheap Wireless 802.11b Bridging More

Cheap Wireless 802.11b Bridging

Comments Filter:

  • Windows-only? (Score:1, Interesting)

    by chrisvdp74656 ( 448900 )
    The WAP11 comes with a USB connection for configuration using a Windows-only application.

    Is an official Linux version planned? Or will it be left to the OSS community to write one for themselves?

    Apart from that, sounds cool.
    • Man, what ever happened to the Linux "roll your own" hacker ethic? Where's the killer advantage of Linux if it still needs the active involvement and development resources of big corporations? I mean, I'm the furthest thing in the world from a coder, but there are lots of people out there who could implement this in their sleep...

      • The main reason for the 'roll your own' ethic is that, back in the day, no sane hardware manufacturer would write a driver for a hobbyist OS with ~20K users. Nowadays, with IBM getting behind Linux and seeing it widely deployed as a fast and reliable server, it's reasonable to ask for some hardware support out of the box.

    • The WAP11 comes with a USB connection for configuration using a Windows-only application.

      Is an official Linux version planned? Or will it be left to the OSS community to write one for themselves?



      The WAP11 can also be configured via its Ethernet port with SNMP. See ftp://ftp.linksys.com/pdf/wap11ug.pdf :

      Linux/Unix Users: Though these operating systems are not supported by Linksys, the Wireless Access Point should work under these operating systems using SNMP. A configurable Management Information Base file (MIB) named AT7C510.MIB can be found on the Setup Utility CD.

    • Their Wireless Cale/DSL routers offer web-based configuration, although more pricey and overkill for this sort of project.

  • Question? (Score:1, Interesting)

    by Justen ( 517232 )
    So how feasible would it be to blanket a city, say the size of Chicago, with wireless Internet, a la Metricom, only using 802.11b? Is it possible? Can you do it with a handful of hubs and some major antennas? (How about wiring the Sears Tower, eh? Hehe.)

    I've seen the site that plans to do it on a user-informal level... But would this be a feasible business or non-profit venture? (Say UofC and UIatC in Chicago, or AUofP and UofPatS in Paris...?)

    Sorry for my lack of knowledge on the issue. I just make sure my WiFi station and cards are working and go on...

    Thanks for any answers.

    jrbd
    • How about a P2P network with IPv6 all over the city? :)
    • SeattleWireless [seattlewireless.net] and PersonalTelco [personaltelco.net] are already doing it. Go read. :)
    • This is pretty funny, as the Sears Tower is already running a service! But not free. Sprint Broadband is using the licensed ITS/MDS band (licensed via nonprofit/ed institutions) in the 2.5 GHz band to offer line-of-sight at megabits per second from the Sears Tower.


      The issue with 2.4 GHz is licensing. You probably couldn't offer enough power with a device and antenna that complies to FCC Part 15 regulations that would have enough range, support enough users, etc. Still, interesting idea!

  • if you are lucky enough to have an antenna within your computer you could very well set it up as a wireless hub.

    I set up a wireless network consisting of 1 Win2k Athlon, a Dual g4 X tower, and an X TiBook for arround $350.

    It's been working fine (as long as the 'puter with the software router doesn't go down). I saved about 300 bucks ;).
  • It would be great to see these wireless
    nodes forwarding packets via each other,
    from node to node, not just from node to
    net. Let's say your and my net connections
    went down for an hour. In the meantime, my
    packets could hop from my node, to yours,
    to someone else's, and so on until they
    reached a node that had connectivity. By
    effectively combining several ISPs,
    reliability and bandwidth would be boosted.
    We'd still need ISPs, but we'd need them a
    lot less, so they might be inclined to offer
    better prices and products.
    • It really sounds good, but connectivity isn't that easy to just up and "share". Unless you and your buddies intend to get a CIDR block, an ASN, and convince your cable and DSL provicers to let you BGP peer with them, I wouldn't count on this idea working at all.
      • You make good points, and there are also security concerns, but remember the early internet had its naysayers, too. One thing they said would never fly was the idea of computer installations actually sharing bandwidth with each other, forwarding packets for each other. Who would ever want someone else's traffic burdening their resources, they asked.

        BTW I didn't have in mind buddies, but anyone in range. Some of the security and care-of-forwarding issues that are being worked on by the IETF for mobile ad-hoc networks could be applied here.

      • Nice Internet buzzwords. Too bad you aparantly don't understand any of them.

        You don't need to run BGP, or a block of public addresses. All of the cheap Cable/DSL routers do PAT and could handle this quite fine as long as you don't have too many nodes.

        • If you read the actual comment I replied to, it implied that if your primary connection goes down, you'd somehow be able to fall back on another one in your wireless network. That is not entirely correct. NAT will only work properly if everyone on the network is using only one connection to the Internet.

          Assuming everything is up and operating, each local node is going to want to use its own egress point through their Cable/DSL provider.

          Now, if one of the cable modems dies, it is entirely feasible for the NAT box (a *nix box most likely) to determine that the modem is dead, stop translating, and push the new connections out one of the other egress points, requiring a reasonable amount of intelligents to determine if a fault is present. Internally, you'd want to use some kind of IGP for advertising default routes so when one dies, another route takes over.

          Unfortunately, all connections that were running to/from that public IP address are going to drop and will need to be re-established. Granted, this is better than a loss of service for several hours. When service is restored back to the new point, the translations will likely go to hell again as the preferred egress point is once again put back into use.

          It gets more complex if there's a problem further than the immediate node or DSLAM, in which case, the NAT box may never fail over to another one of the links. For example, if I can get to my default gateway on my local node, but not beyond that...and my NAT box isn't set to detect that kind of failure or doesn't think there's a problem, then I still have a service outage. The same issue applies to DSLAMs.

          Use of a CIDR block would give everyone in the wireless domain a unique public IP for each front-end and wireless address in the network. Backend and private LANs could still be NAT'd on a site-by-site basis. A CIDR block would probably be required with multiple providers, since exporting more-specific prefixes of a larger aggregate block could cause severe traffic shifting problems.

          If each of the egress points were to run BGP with their respective gateways, or eBGP multihop to a more reliable headend or provider core router, the problem of network reachability would be much less.

          Take, for instance, an outage between the cable node and the headend. If a BGP session were established with the headend, all routing between the headend and the node would cease in an outage and as such, and traffic would not traverse the cable link since there would be almost no routes coming in over the failed link. Even better, a BGP session with the headend would quit entirely and no routes would be received for that link at all.

          Since all the BGP routers inside the wireless network would either need to be fully meshed or run as a series of router reflectors and route reflector clients, there would still be acceptable internal routing if there were indeed a failure at one or more egress points.

          That's why an implementation using BGP and CIDR blocks would be the preferrable way to handle it.
    • Yes, indeed, this would be excellent to have, however it requires the use of a routing algorithm to determine where to send packets. I'm sure the Linksys folks are working on this right now.
      -russ


    • You're right - and it exists. Routing protocols that would make such things work exists for so-called MANETS (Mobile Ad-hoc NETworks), being developed by the IETF.


      Working in this area myself, I'd like to point to [ietf.org]
      http://www.ietf.org/html.charters/manet-charter. ht ml, which is the IETF-working group dealing with MANET's.


      Now, for the shameless plug: A link to the OLSR routing protocol for MANET's [ietf.org], which is showing promising results. Implementations (downloadable, with sourcecode etc. of the routing deamon) are available (drop voop@cs.auc.dk an email if interrested in the code - the www-server is currently not responding).

  • Gah... (Score:2, Interesting)

    by Sokie ( 60732 )
    Why must people keep coming out with things that make me feel like I wasted the ~$600 I spent building a wireless to wired router/bridge out of one of these [allwell.tv] (the 1030N if you are wondering). Granted I could have saved some money if I had be brave enough to try to use a DiskOnChip instead of buying a 2.5" HDD...

    Oh well, when I built it I still think it was cheaper than any of the other available solutions. Except of course for the P-133 box it replaced...but I wanted something that would sit on a shelf in the garage very unobtrusively.

    BTW: here is the box [wwc.edu] sans hard drive.
  • "Using higher-gain antennas than the ones shipped with the WAP11 ..."

    That's a nice feature, but it is not mentioned in the article.

    Who knows more about this ?
    • Irkutsk region (Siberia, Russia) record for 2.5 mbit/s Wavelan is 47 kilometers, using hi-gain parabolic antennae and the direct line of sight. But the use may contradict US FCC regulations that limit the antenna gain. If you want to risk conflict with FCC and cannot buy the antennae you can use the satellite dishes and make the exciter (don't know precise term for a little antenna in focal point) yourself. If you want to make the hi-gain OMNI you can disassemble the hi-gain cellular or CT2 antenna and make the scaled copy. Remember: Cable losses are terrible and you should place your radio near the antenna.
    • Go search the cisco [cisco.com] web site for aironet and antenna, they have a few other antenna types, including a nice +21dB parabola not available in Europe.

      Because 802.11b devices are being sold to consumers, they are required to have "non-standard" connectors not readily available on the market so people can NOT modify the antennas to boost range. That is a requirement in the U.S. and Europe, so all 802.11b manufacturers use Reverse-TNC or Reverse-SMA connectors.

      The article had it wrong when it said the units had standard connectors. Clearly the author just bought two boxes and hooked them up and they worked, just like the TFM says. This article didn't deserve a /. listing, but in these last few hot summer days, the news is pretty thin.

      the AC
      • There are people, in Europe (and no doubt, elsewhere) building their own high-gain antennas quite successfully. Work out dirt cheap, too.

        Adapters are available to go from non-standard connectors to something a bit more garden variety, BTW.

        ...j
      • You can get commercially available adapters to do the conversion. You end up with a slightly messy daisy-chain, but it works.
        I set a similar system up to cross a road a few months ago. I user Intel APs and the cisco antennas.
        One piece of advice: don't use Intel APs. They fail @%$#@%$ constantly. I've had to get twice as many as I need to cover while they go for warranty (but at least the warranty's solid :)
      • Actually, the Linksys units come with simple, standard connectors that folks sell antennas to attach to for cheap or expensive, as you prefer. Try the lists.bawug.org archives - they have a lot of posts on companies that sell antennas, connectors, etc. Some requires soldering, but a lot are just order the right part and attach. (The Lucent/Agere cards from Orinoco, for instance, have a cover on the female plug - you remove it and plug an antenna right in. Orinoco sells its own, or you can go third party.)

    • Here [infoworld.com] is an article at Infoworld [infoworld.com] about an 802.11b freenet deployment effort. There is mention of a home-brew antenna using a Pringles [pringles.com] can to boost the gain. No picture of this cool hack, though.

  • Legal issues (Score:4, Informative)

    by mwillems ( 266506 ) on Monday August 27, 2001 @05:35AM (#2220542) Homepage
    Careful though.

    a) By modifying equipment you may be breaking FCC rules (USA) or your local rules.

    Additionally, in the UK, 802.11b is NOT apprived for commercial use. I spoke at length with the UK government Radio Agency [radio.gov.uk] last week to establish this (my company use 802.11b to connect remote advertising screens, but not in the UK where this is forbidden.)

    Michael
    • > a) By modifying equipment you may be breaking FCC rules (USA) or your local rules.



      Have you seen one of these access points? Putting a high gain antenna is hardly breaking any FCC rules esp since the Linksys access point has two fairly standard antenna connections.


      Unfortunatly, no hacking is required.


      Chase

  • Additionally, in the UK, 802.11b is NOT apprived for commercial use

    The same is valid for Austria and Germany. I hope that worldwide standards will soon be developed so that we can avoid this "inconvenience"
    • Don't confuse commercial use with private use. I still think there are questions out about this in the U.S., too. But in most countries in the world, the 2.4 GHz band has been harmonized enough to clear 1 or more 22 MHz (full bandwidth) channels necessary to run 802.11b. You can use them for business purposes, but you cannot deploy and resell network access, as MobileStar, Wayport, and others are doing in the U.S.
  • Anybody knows a wireless USB device working with
    Linux ? I couldn't find any :-(
  • 802.11b IS approved in Germany and Austria, i just read this here [press1.de]
  • I can't wait to see this done using 802.11a devices (when they arrive). :)

    Last mile problem? No problem at all! :)

    [For those who don't know, 802.11a is just like 802.11b, except at over 50 megabits per second.]

  • Broken... (Score:3, Insightful)

    by meisenst ( 104896 ) on Monday August 27, 2001 @07:25AM (#2220651) Homepage
    Unfortunately, like other 802.11b solutions, these use WEP, which is inherently broken... if I were linking my business' two campuses, I wouldn't really want Joe, Bob and Mary (who all happen to have line of sight) sniffing all of my network traffic with their perfectly simple store-bought solutions. :-)

    meisenst
    • The only real solution for the problem WEP tries to solve is to use end-to-end encryption, e.g. ssh or ssl.
      -russ
    • Yeah, the author mentions that. I wonder if FreeS/wan [freeswan.org] might be a good solution. Just throw a coupla out-of-date machines in front of each WAP box... FreeS/wan offers the ability to plug in the encryption method du jour and doesn't require any reconfiguration of your client machines...

      Food for thought...

      • I wonder if FreeS/wan might be a good solution.

        Yes, it is a good solution. I can neither confirm nor deny having taken part in just such an implementation. :-)

  • Detalils? (Score:2, Insightful)

    by Overphiend ( 227888 )
    Is it just me or is this article completely lacking in any technical details at all. Basically there's nothing noted there that isn't in the users manual. It's even lacking much of what's in the manual. And it mentions a better antenna you can connect to the WAP11, but does it mention any details about the antenna? No. I've been looking at the antenna's on the market and I've found the HyperGain HG2410U 10dB Omnidirectional Antenna, however I still cant figure out which of the like 10 different connectors will go into the back of the WAP11.

    • Re:Detalils? (Score:2, Informative)

      by oldave ( 160729 )
      Reverse TNC Male

      And you can use any antenna that's built for 2.4 GHz. Unity gain, 5.5 dBi (what I'm using to link 2 condo complexes), etc.

      Use LMR-400 (or better - better is left as an exercise for the reader) cable, reverse TNC Male connector for the back of the Linksys... N-type female (usually) on the antenna end of the cable.

      Calculate your system gains/losses at this site:

      http://www.dct.com/~multiplx/wireless/wireless.mai n.cgi?GoButton=Go+Now%21 [dct.com]

      A word of caution - don't mess with the connectors yourself unless you know what you're doing. A stray strand of braid touching the center conductor will blow you out of the water. Also, to minimize loss (we're only talking 50 milliwatts here - not much power), be sure you use quality, solder-type connectors, and only the ones at each end... "measure twice, cut once" is an invaluable piece of advice when installing your transmission cable.

      Also be careful of short coax runs and 24 dBi gain antennae - don't want to run afoul of the FCC.

      One other comment - don't use channel 6 (the default in the WAP 11).
  • I tried looking to see if there's some way to use a second wireless access point as basically a repeater (so I could extend my wireless network throughout the house). I was hoping for something that'd just look at the packets coming in and send them out to and from the previous access point.

    Even called Cisco and they said this couldn't be done. Anyone know differently?
  • My best friend lives about 10 miles away (not line of sight :( ) and he has a DSL connection. My parents live approx. between us. I need something wireless that would allow me to use his DSL line and we could use my parents house as a repeater.

    So, does anyone know of any 5 mile, non line-of-sight, bridgeable, wireless solutions? Or am I asking too much? :)

  • $199 Cheap? Make it yourself. (Score:4, Informative)

    by steevo.com ( 312621 ) on Monday August 27, 2001 @08:45AM (#2220823)
    Why not turn bridging on in the kernel and bridge to an internally connected 802.11b card.

    I used to work for (company unnamed, but you have heard of them) a place that developed a stand-alone product where the "bridged mode" is done exactly this way. (It was still in development when I left, and I don't believe it is yet on the market.) Bridging performance was about the same as other 802.11b access point devices.

    While the platform was a bit differant (StongARM and mini-PCI cards) there is no reason you couldn't do the same with a P166 sitting in the closet, a card bus controller and a cheap 802.11 card.

    • Why not turn bridging on in the kernel and bridge to an internally connected 802.11b card.

      Because I bought Orinoco Wavelan cards and the bridging firmware is sent to the card by the access point at every boot up. :-(

      Mind you, I've got a logic analyzer with enough lines to handle a 16-bit PCMCIA bus, I just have to beg/borrow/steal an access point and spend the next few weeks untangling the init code to see what they send so I can tell Linux to do the same. :-)

      Are all cards that use the same chipset as the WaveLan cards (Hermes 2 I think?) incapable of bridging without the firmware? If not, I'll just buy a cheap one and use it instead, or see if I can't do a dump of the firmware and muck with sending it to the Wavelan. :-)

    • If you can build a Linux box and stick a generic wireless card in it for less than $185 (the Linksys including rebate) and a couple hours of your time, then it's certainly a good idea. I know that BAWUG (www.bawug.org) is working on a platform reference for a 486-based, low-power system that would be a Web server, access point, authentication server, and lots else.
      • Couldn't do it for $200 if all the parts were new. (At least using common PC hardware. You could build an embedded device in volume for that, but that's not my point.) Most IT folks that I know have lots of old hardware hanging out with nothing to do.

        Couple of hours. Not the for the first time, at least it wasn't for me the first time I did something similar. Certainly, there is a learning curve. However, once complete, it would take very little time to duplicate this effort. Factor the time over multiple units, and this is no longer an issue.

        Sure you can spend $185 for a solution that works almost out of the box, but it's more fun to MacGyver something together using obsolete equipment around you and a minimal amount of cash.
      • The Linksys PC Card is terrible. Ten unobstructed feet from my WAP (a Linksys WAP 11, which I'm quite happy with) and the signal strength was barely 50%. Completely unusable 30 feet away thru a wall. Switching to an Orinoco Silver cured these problems. And I'm not the only one who had this difficulty; others on the alt.internet.wireless newsgroup report the same poor reception problem.
    • You have to understand about signal loss in the antenna runs. If your computer is not near where you need to put the antenna, then you could lose too much signal just getting to the antenna. Not everyone would take this into account and wonder why they can't even get a signal from one side to the other.


      Using an access point means that you have a cheap computer than you can put almost anywhere and will (hopefully) just work. You can put it in the attic, just under the antenna. You can, theoretically, put it under the eaves of the roof (though they are usually not designed for outdoor use, so be careful).

  • There's a story [free2air.org] (very light on details) with some pictures of my node sitting 450 meters away from it's wireless uplink.

    You'll notice the use of an empty cider can to mount the Lucent Range Extender on. This was actually vitally important. Between the two sites was a kebab shop (seriously) that was just breaking line of site, and with the various combinations of wireless kit, we were right at the end limits of getting a signal. We messed around for a few days trying various things, and, eventually, over a can of Strongbow I realised raising the antenna those 5 or 6 inches higher might work. And the rest is history.

    That site has moved now, and is much closer to the primary site,so the feat is not nearly as impressive. But you geeks can, uh, geek out at knowing that the website below comes to you [spooky voice]through the air[/spooky voice].

    ...j
    • Oh, and if you're feeling like being geeky about 802.11b and related things, make sure to look at www.free2air.org [free2air.org] for some good detailed stuff about wireless security. Airsnort, war driving, and all that other stuff you kids seem to love.

      The guy who does the site is in Amsterdam this weekend doing some wireless stuff. Wait 'til you see pictures of his latest project. But i'll let him explain that when he's ready...

      ...j
  • I've just used the Linksys firmware to upgrade
    my netgear me102 access point and used the Linksys
    SMNP utility to configure it.

    In fact I think it should work on any PRISIM chip
    based 802.11b access point.

    One thing I noticed though. After upgrading the firmware I had to unplug the ME102 and plug it back in brfore I could access it again.

    • Geeks beware: as of this writing the Linux drivers for the Linksys WPC11 are IMHO unusable for setting up a simple home firewall/router. This is the cheapest Linksys 802.11b card out and it's the only one that Fry's carries. The WPC11 requires an experimental Linux kernel plus patches plus editing some constants at the beginning of a header file and recompiling a driver to get them to work. Apparently the WPC11 is basically just a glue card that makes the PCMCIA core visible on the PCI bus. See here [linux-wlan.org] for more info on getting the Prism chipset working with Linux and problems with the WPC11. If you want to build a working router with the WPC11 in less than a day with a spare PC, you'll have to swallow your open-source pride and install a copy of Windows 98 and WinProxy. I did so last week and got it working in less than an hour.
  • "Using higher-gain antennas than the ones shipped with the WAP11 and/or tuning line-of-sight access..."

    That's fine and dandy to raise the gain on the bridge or router or whatever but the gain on the client side will also have to be raised or you won't see any increase in distance in your wireless networks.

    • That's not necessarily true.

      To put it in the simplest terms - increasing the gain of the antenna effects not only the transmit range (more power out) but the reception range (more received power delivered to the receiver)

      Of course, increasing gain also narrows the beamwidth - which means that alignment of the antennas becomes increasingly critical as the antenna gain goes up.
  • It is hardly theisable to setup a wireless network with lynksys with wireless internet. It is however entire possible to do this with other manufactures gear. For example in Minnesota MANY cities have a partial deployment of wireless internet, and a few have FULL deploymen, the city of mankato (the home of MSU mankato state university) has 100% of its city wired. In Minnesota, you can drive from Midstate, to northern iowa without ever getting off the internet. ericc@xtratyme.com
  • But higher-gain antennas are illegal modifications to that sort of device. The rationale is if 12 of you in the same apartment decide to wireless wan yourselves to 12 separate friends across the street with you high-gain antennas or amplifiers, at least one pair of you, and possibly more, will experience service degredation. Just like the CB syndrome, where people using legal equipment got stomped on by the many using illegally modified equipment. It gotten to the point where the FCC only brings enforcement against CB ops in cases of interference to other services, and even then only against stationary operators, 'cause you can't bust everybody with 18 wheels, a brick and a firestick.
    • Illegal modifications?

      By whose definition?

      CFR 47, 15.247 defines maximum peak power, among other things (such as the frequency hopping intervals, occupied bandwidth, etc).

      FCC rules do NOT specify that you must use the antennas supplied with the system.

      Quite specifically, in fact, the rules leave the choice of antenna to the user... within specified limits.

      May I suggest that you read the actual rules before making such a general comment? Here's the link:

      http://frwebgate.access.gpo.gov/cgi-bin/get-cfr.cg i?TITLE=47&PART=15&SECTION=247&YEAR=2000&TYPE=TEXT [gpo.gov]

      • I bow to your pedanticism, and congratulate you on reading the CFR from end to end as you're implying. The article referred to said something like "hang a higher-gain antenna off the back" which implies an illegal modification to me. As long as you read CFR 47, do you remember reading in there somewhere in part 15 about modification to equipment, including class B digital equipment, and what the end-user/consumer can and can't do? Oh. And even in your cite, are you confident that if you rig up what's described in that article and the FCC comes knocking that you can demonstrate that you're employing transmitting antennas with directional gain less than 6dbi, or greater than 6 dBi and ensuring that the maximum peak output power of the intentional radiator is reduced by 1 dB for every 3 dB that the directional gain of the antenna exceeds 6 dBi? Oh.

        May I suggest you re-read 15.209? Are you quite positive that if they start measuring microvolts of RF around your equipment to apply 15.209 you're still compliant? Oh. And even if you say "but I bought this antenna from Lucent" (or whoever) to pass the buck to whoever certified your equipment to be part 15 compliant will they say "oh, well it must be okay! Bye!" Oh.

        My guess is, if somebody's cell phones or electron microscope are flaking out in your neighborhood and the fcc and proxies go sniffing around and notice 802.11 equipment with yagis hanging off the flagpole and decide you're a good scapegoat, at best you'll just get your 802.11 equipment confiscated, at worst they'll take all your electronics, search your house, call in other agencies, and fine you, and you can cite the CFR, Constitution, Blackstone's Commentaries and the menu at McDonalds and it won't help.

        Sometimes a general statement is more accurate than selective citations, implications of ignorance, and rhetorical "by whose definition?" sort of questions. In any event, you should not try to bolster your arguments with cites that only support your somewhat narrow and amateur (pun intended) armchair lawyer interpretations unless you're damn sure you're dealing with somebody totally ignorant. I grant you it works with several folks on here, but I don't roll over.

        Might I suggest you make sure know what you're talking about and don't just do a cfr search for antenna before you suggest that I read the actual rules?
        • Actually, I can demonstrate RF field strength quite handily, thanks.

          Not to turn this into a flamefest, but your repetition of 802.11 makes it sound like there are special rules for that protocol. There are not. The rules in question cover the IMS bands.

          Please be sure that I can demonstrate that my installations are part 15 compliant. Granted, many couldn't, and can be led astray by general statements like "hang an antenna off the back."

          Since the WAP-11 is spec'd to do 54mw out, with 10 feet of LMR-400 into a 24dBi gain antenna, you're gonna be something over 5 watts effective... way illegal.

          In the one situation where I've "hung antennas off the back" of a WAP-11, I'm getting something like 90mw effective... at 30 feet AGL.

  • Exactly what is line of site? If I wanted to use this method to bridge two wireless networks together would I acctually have to be able to see both antenna's when I stood by the other, meaning no tree's or anything could be obstructing the view?
    • Line-of-sight means exactly what it says. Yes, you must be able to see the other site.

      (yeah, I know all about the radio horizon vs. visual horizon, and Freznel zones, etc. - but for most purposes, with these devices, radio horizon isn't going to be an issue)

      Trees, buildings, mountains cannot obstruct the view. The frequency ranges used do not pass through objects very well at all.

      If the two sites are close enough, with enough gain (remember, gotta keep it legal, though), some amount of trees, walls, etc, may work.
    • Actually, you need better than optical line of sight. The radio waves need to have a clear space the shape of a lens. Otherwise you get multipath distortion. So yes, count on needing line of sight.
      -russ
    • I'm no expert on radio/wireless technology. However, I can tell you about my setup at home:

      One Cisco Aironet 350, sitting on top of my 6ft 4 port rack. I can't see it from my laptop, but it still works. I can go into another room (with 2 cinder block walls between) and it works fine. I can go out into the yard, it works fine. I can sit in my car and drive 4 houses down and it still works (so, 3-4 cinderblock walls, a wood fence, and my car to pass through). Much past there and the link starts failing pretty fast, but then I'm getting outside of the distance specs anyway.

      I do know that metal and certain thicknesses of materials affect it, direct line of sight isn't needed from the client to the access-point.

      Now, perhaps this story is focusing specifically on long-range, and for that, I'm sure line of sight plays a much bigger role in affecting the signal, and I'm sure the type of antanae matters as well.
  • If I wanted to end up providing tech support to my neighbors. I can think of no less effective use of my time.
  • I've been looking into putting wireless between my friends apartment and mine. We live in the same apartment complex and are about a half mile away. But in between us is several walls and a bunch of trees. Is there a high gain antenna I could slap on one of these puppies that would burn through something like that?
  • This isn't exactly new, but the article I wrote, which just went up on Friday...

    and it went down at:

    Posted by Hemos on Monday August 27, @03:38AM

    Sure fancy boy you can do all that reading and writing but can you do mirroring?

  • You could also use Windows XP to do this with two wireless cards in Ad-Hoc mode (dont really need an AP in a small office/home/apt.) as XP will bridge the wired/wireless networks together.
  • would be a wireless access point with a built in switch. Do these things exist for a cheap price?

    The problem is that if my cable modem is on the opposite side of the house where my lan is, then I'd have to buy wireless cards for all my pc's. Or I'd have to buy at least one card and use a computer as a router. bleh.

    I'd rather have an access point with a built in switch/hub that would allow me to cascade it into a remote lan/switch. If these things do exist, are they cheap?

    • BEFSR41 [buy.com] - Router + 4-port switch - $99.95 ($84.95 after rebate)

      WAP11 [buy.com] - Wireless AP - $195.95 (180.95 after rebate)

      BEFW11S4 [buy.com] - Router + 4-port switch + Wireless AP - $224.95 ($199 after rebate)

      I don't know why someone would by the WAP-only model since it's only $20 more (after rebate) for the version with a Router+4-port switch. If you had the BEFSR41, just sell it to a buddy for $50, buy the BEFW11S4 for $199 (after rebate), and you end up saving $30 (plus your friend gets a good deal on the BEFSR41... Unless you've got some special router already and don't want to sell it).

      • I don't know why someone would by the WAP-only mode



        Somebody contradict me, but as far as I understand it, only the WAP11 is designed to take the bridging firmware upgrade. Likewise, once you've switched to bridge mode, you can't use it as a plain access point, too.



        I'm guessing the original purpose was to have two price points, when they were $50 or $75 apart in price.

  • xset up to boxes with your favorite os, pop in a wired and wireless nic in each, put the wired ends on there own networks (192.168.1.xxx and 192.168.2.xxx) and use a diffeent set for the wirelesss portion (10.15.3.xxx). set up each box to route, and run the wireless segment using encrypion L2tp comes to mind..
    and enjoy.

    not the cheapest, but secure none the less, and you can allways stick a crisco logo on em :)

Slashdot Top Deals

The rule on staying alive as a program manager is to give 'em a number or give 'em a date, but never give 'em both at once.

Close