Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 


Forgot your password?
Close
typodupeerror
×
Hardware

802.11 RF Amp 229

Posted by timothy from the from-hobby-bench-to-compusa dept.
MikeLRoy writes "Linksys has announced a signal amp, to be available soon, for their wireless ap's/routers. While many people have been using commercial rf amps hacked onto their ap's, linksys now has a commercial solution!" I wonder when ISPs are going to stop soft-pedaling the anti-NATing provisions in their terms of service.
This discussion has been archived. No new comments can be posted.

802.11 RF Amp More

802.11 RF Amp

Comments Filter:

  • Hmm cool. (Score:4, Funny)

    by Anonymous Coward on Sunday December 22, 2002 @06:01PM (#4941923)
    While many people have been using commercial rf amps hacked onto their ap's, linksys now has a commercial solution!

    Now if they could just come up with a commercial solution, we'd be all set.

    • Re:Hmm cool. (Score:1, Funny)

      by Anonymous Coward
      forget it, I'm sticking with the commercial solution

      (usually it's better than the commercial options)

      then again, I haven't seen what the commercial vendors are offering!

  • One way. (Score:1, Insightful)

    by Trusty Penfold ( 615679 )


    To improve the range of a wireless connection you would need an amplifier at both ends. With this large box attached to the access point, the AP can transmit packets further ... but this is useless if you laptop, say, hasn't got the power to transmit packets back. This thing just introduces a zone of one-way communication around the inner zone of 2-way communication.

    • Re:One way. (Score:3, Informative)

      by Anonymous Coward
      Actually, this device increases the RX sensitivity and power as well as that of the TX, thus giving a greater range regardless of what you have in your laptop.

    • Re:One way. (Score:4, Informative)

      by ZeLonewolf ( 197271 ) on Sunday December 22, 2002 @06:07PM (#4941952) Homepage
      To improve the range of a wireless connection you would need an amplifier at both ends. With this large box attached to the access point, the AP can transmit packets further ... but this is useless if you laptop, say, hasn't got the power to transmit packets back. This thing just introduces a zone of one-way communication around the inner zone of 2-way communication.


      Erm, no... it's a signal amp... so if you stick it halfway between a laptop and a wireless AP, it will amplify the signal from both. It's just like using a repeater in a cabled network.

      • Re:One way. (Score:2, Informative)

        by Andorion ( 526481 )
        Actually, like puzzled (12525) posted: "FYI half of the reason to deploy an amp is for the LNA (low noise amplifier) effect - besides boosting output they pump up the return signal by 10 - 14 dB."

        Give him the karma. And mods, why the hell are you modding up this post's parents, and it's parent's parent?

        The amp doesn't go halfway between anything, and you don't need a similar device on the other end.

        -Berj
      • But its not stuck between the laptop and the AP. Its stuck ON the AP. If the incoming signal can be amplified then its strong enough for the AP to pick it up in the first place.

      • Re:One way. (Score:3, Informative)

        by div_2n ( 525075 )
        I think you are confused as to what an amp is and what it does and doesn't do.

        An amplifier takes raw signal (including noise) and amplifies the power on it in the direction it is set to amplify. There are bidirectional amplifiers to do it in both directions.

        Amps are best placed as close to the antenna as possible to reduce loss and noise. As a signal travels across cable, noise and loss increase. Thus, a short "jumper" cable is ideal to connect the antenna to the amp.

        Putting an amp on one side of the communication pathway will improve the signal and perhaps the distance a bit. The biggest improvement will be in the quality of the wireless link at the far reaches of the original signal.

        To greatly improve distance, it is absolutely necessary to amplify on both ends. The end goal is to improve your signal to noise ratio. The weaker your signal from the other end, the worse that ratio becomes.
        • What you're ignoring here is that the receiver has a limit on its sensitivity - So that even if it is receiving a signal with a good SNR, if that signal is too weak, it will be below the noise floor of the receiver itself.

          Adding a preamp boosts that signal. Yes, it amplifies the noise, but it will bring the signal above the receivers' internal noise floor. The most important factor in the noise performance of a receiver is the noise figure of the very first gain/loss stage after the receive antenna. Put in a low-noise gain stage here and the NF of the entire system drops. (This is why satellite receivers have a preamp at the dish, not at the receiver at the other end of the coax, and why broadcast TV amps should be at the antenna, BEFORE the coax run)

          WLAN receivers (especially budget ones like Linksys, D-Link, and almost any mainstream Prism2/2.5/3 implementation) usually don't have the best receive sensitivity because a good preamp at 2.4 GHz costs $$$. This preamp is a cheap way around that problem.

          If you want a good example of why a preamp will help you - Use Kismet (a passive receive-only monitor) with:
          a) A Prism-based card
          b) An Orinoco card
          c) A Cisco card

          You'll notice that given almost identical antenna designs and identical signals at the receive location, the Prism-based card has horrible receive sensitivity, the Orinoco is INCREDIBLY good (compared to the Prism), and the Cisco is even better. (Not as much improvement as Orinoco vs. Prism, but still noticeable)

          Note: High-end Prism-based cards like the Demarctech ReliaWave are exceptions. The Reliawave beats Orinocos and even Ciscos I believe.

          So you can get a LOT of receive performance improvement with just a transmit power amp/receive preamp on one end. Especially when the receiver in the AP isn't particularly hot. (I think most APs are better than cards, but at the card end, you'll see a lot more benefits from adding a preamp to a Prism based card than to an Orinoco or Cisco card, since the Prism has the worst receiver and as I mentioned before, the very first gain stage is the most important.)

    • Re:One way. (Score:3, Insightful)

      by balamw ( 552275 )
      Actually, I would presume that this thing has LNA on the receive side, and maybe one for transmit. Many applications like this where the mobile transmitter is small, has low power consumption requirements and inefficient antenna design are reverse link limited, and adding amplification (gain) on the receive side alone can increase range of the overall system. Since the transmitter is stationary and plugged in power is usually not much of a consideration for the transmit side of an access point.

      Balam

    • When you broadcast at higher power, generally you would use a comparable signal amp on on incoming signal. So your underpowered laptop signal is greatly magnified (errors and all) after it is received at the hub. Larger distances then translate into lower available bandwidth as opposed to no connection. This is pushing and pulling signal from one end instead of pushing from both.

      Get yours here! [hazardfactory.org]

      • I doubt that a laptop needs a huge ammount of upload bandwidth anyway. I don't see anyone running servers off of laptops accross a network. So even though the laptop will have errors in the signal it will still only be a little slower than the download speed which is what a laptop is used for (checking email, browsing webpages). Personally i would like to purchase something like this as i have been thinking about making my own but that can introduce my own errors anyway.
        • > I don't see anyone running servers off of laptops
          > accross a network.

          Funny... I do just that. I have a lot of old
          laps that aren't really usable as interface machines
          anymore, but have cardbus 100tx nics, and use a lot
          less power and run a lot quieter than some midi-tower
          would, so i have them in my closet doing nat, serving
          web pages, serving mail, etc.

          In a professional environment, they'd be replaced
          with rackmount systems, but as a good way to be
          green, recycle the hardware, and factor my
          infrastructure for maintainability, old laps as
          servers makes a lot of good sense.

          I'm talking home use here, of course.
          • Ok i knew by saying nobody does it the people who would were goign to speak up but i still believe the same thing. Laptops are generally not used as servers. Your laptops are not running over wireless are they? I can see using ANY old computer for servers in the closet but i cannot imaging a bunch of wireless capable laptops running apache and whatnot.
            • Given that my NAT box is a laptop, rather than
              spending $100 on a WAP, I just used a PCMCIA card
              with an antenna, so, yeah, in this case, one is
              wireless -- although only the p2p services running
              on my actual, physiological, lap are being served
              through a wireless link.

    • Re:One way. (Score:2, Informative)

      by Anonymous Coward
      That's not entirely true. If the signal going FROM the hub to the laptop are strengthened, the data transmission would definately be improved, but in most cases, one can also improve the "reception" capability of the hub by boosting the recieving antenna gain. But by the same token, these systems run full duplex (transmit and recieve at the same time), so if the transmitter is boosted, it COULD concievably de-sensitize the reciever.

      Not sure of the exact machanisms IF the hub actually transmits and recieves at exactly the same time, but now that I think about it, perhaps not (being digital - it COULD be simplex).

      If you've even been near a powerful FM or even a TV transmitter, and try to use an FM radio, you would be hard pressed to pick up anything else BUT the FM transmitter.

      RF can get into ANYTHING... as a 1st class FCC license holder, and former Chief Engineer of a 25KW FN station, I know only too well what RF can do. Ask your food, when you take it out of the Microwave oven.

      • But by the same token, these systems run full duplex (transmit and recieve at the same time)

        No, 802.11b is not full duplex. You transmit and receive on the same channel in the 2.4 GHz spectrum. For example, if you are using channel 6 within the 2.4 GHz spectrum (one of three non-overlapping channels), you are Tx and Rx on the 2.437 GHZ frequency.

        In many ways, 802.11b Access Points give you similar performance to what you would find in a 10BaseT hub, which were very common in Enterprise networks 5-10 years ago. Some of the same caveats apply... At a certain point, adding additional hosts to the same collision domain really decreases the available bandwitdth to each host.
        • In many ways, 802.11b Access Points give you similar performance to what you would find in a 10BaseT hub

          It is actually a bit worse. Bad signal levels, noise, multipath and other issues conspire to make the throughput of the average "11Mbps" 802.11b network lower than you would expect on a regular 10Mbps 10baseT hub with a similar number of clients.

          Clients scale back to 1, 2 or 5.5 Mbps if the signal quality is not sufficient for 11. A client sending 1K of data at 1Mbps use the same air-time as a client sending 11K at 11Mbps.

          On a shared hub, all clients can 'see' each other, so when two nodes transmit at the same time they detect a collission. On wireless, you can have hidden nodes (both clients can hear the AP, but they can't hear eachother) so you can get collissions but the clients don't know that they collided. To mitigate this problem, you can revert to an RTS/CTS system (AP tells nodes when they can send) but at the cost of lower throughput.

          2.4GHz is also becoming crowded. Bluetooth, DECT phones and other equipment also use the same frequencies. This creates more noise and interference.

          Etc..

          When having more than a few clients, expect aggregate thrughput on 802.11B to be closer to 4-5Mbps than 11.
      • Errr... there is no longer a 1st class FCC Commercial License. Mine turned into a "General Class" a couple of decades ago. Second class also turned into a "General Class."

        (for those who ask, this is a kind of license needed to be a technical person on various kinds of FCC stuff, such as two way radios and broadcast stations). The term "engineer" in broadcasting is like the term "engineer" in trains - it doesn't mean you are an engineer. I started as a broadcast "engineer" when I was in high school with a 1st class FCC License (commonly called "First Phone").

      • RF can get into ANYTHING... as a 1st class FCC license holder, and former Chief Engineer of a 25KW FN station, I know only too well what RF can do. Ask your food, when you take it out of the Microwave oven.

        I used to be a radar tech. I used to fondly call the things and open air microwave oven.

        Which brings up the usual questions of frequency range and power. One or two probably are no great shakes, but a Special High Intensity Test would be needed if you had a bunch of them

        ;-)

      • 802.11 is not full-duplex.

        To the user, it may appear full-duplex because it switches between transmit/receive extremely quickly, but it's not full-duplex, just like 10 or 100baseT on a shared hub is not full duplex but without extensive benchmarking in a high-load situation, the user can't tell that it isn't FDX.

        Do a search for N9ZIA - The guy is a bit nuts, but has done a LOT of 802.11 hacking, including some major mods including bidirectional amp ideas for Proxim's 802.11 precursor products. (Not 802.11, but very similar)
  • I've been waiting for a legit solution for so long, I'm so glad they finally decided that it was worthwhile to produce one that is not hacked together

  • Huh? (Score:1, Offtopic)

    by bobdotorg ( 598873 )
    I looked at the link, and was wondering: How the hell is this thing FCC compliant in the U.S.?

    Now we only need Pringles to intruduce a bigger can.

    • Re:Huh? (Score:4, Informative)

      by LostCluster ( 625375 ) on Sunday December 22, 2002 @06:09PM (#4941964)
      Because this device emmits at the FCC max, something most APs do not. Yeah, this thing would rock inside a Pringles can, but it's likely even the worst designed can will put the setup over the limit.
      • The fundamental design flaw. I would rather eat sand than that crap. My dog seemed to agree.
      • I haven't seen any specs, but I have a feeling this doesn't emit at the FCC max.

        In fact, I wouldn't be surprised if it's not anywhere close to the power output of some of the $300-400 range of bidirectional amps - I can't find specs on Linksys' website, but I wouldn't be surprised if this only boosted the signal to 100-200 mW. Legal limit into an isotropic antenna (or is it a dipole?) is 1W.

        Also, FCC regs on the EIRP are a bit odd - There isn't an exact fixed limit on EIRP, i.e. for every 3 dB of antenna gain you add, you don't have to drop transmit power by 3 dB - It's only 1 or 2 dB or power you're required to drop.

        http://www.qsl.net/n9zia/wireless/page12.html has specific info.

        +30 dBm max output at the product's connector
        +36 dBm max EIRP for multipoint connections
        For fixed point-to-point connections, the PEP of the transmitter must be reduced 1 dBm for every 3 dB of antenna gain beyond 6 dB.

        i.e. with a 6 dB antenna, you may run 30 dBm (1W) output power, for an EIRP of 36 dBm. With a 9 dB antenna, you must drop to 29 dBm output, for an EIRP of 38.

        At 20 dBm output (100 mW), you may run 36 dB of gain
        At 23 dBm output (200 mW), you may run 33 dB of gain

        If the max output of this amp is only 20-23 dBm as I suspect, then not even the best Pringles antenna won't be able to push it beyond the legal limit. Only a large high-quality dish will have a chance. (33 dBi is a LOT of gain. The largest parabolic antenna at http://www.fab-corp.com/ only has 24 dBi of gain)

    • I looked at that link and I kinda got a stiffy (Score:1, Funny)

      by Anonymous Coward
      Sick huh? Something about that photo looks like fun.

      Sure hope this is anonymous.

    • Are You Daft? (Score:3, Informative)

      by m1a1 ( 622864 )
      It is just a repeater. It doesn't broadcast at any greater strength than your typical 802.11b access point, all it does is repeat whatever it picks up, thus increasing the range.
      • Why don't I ever have mod points when I need them?

        The device, while so far light on specs, is marketed as an amplifier, not as a repeater. From Linksys' product page - "The Linksys Wireless Signal Booster piggybacks onto your Linksys Wireless Access Point (or Wireless Access Point Router)"

        Repeaters don't "Piggyback" on the AP. They're placed elsewhere.

        Also, if you look at the picture, esp. the enlarged one on buy.com's product page, you'll see that the booster is stacked on top of an AP in the pictures - WITH COAX RUNNING FROM THE AP TO THE AMP. Also confirmed on Linksys' product page - "To install, just stack the Wireless Signal Booster on your Access Point, move the antennas to the Booster, and attach the Booster's twin cables to the Access Point -- no drivers or modifications to your setup are necessary."

  • Anti-WAP? (Score:5, Informative)

    by Motherfucking Shit ( 636021 ) on Sunday December 22, 2002 @06:05PM (#4941942) Journal
    I wonder when ISPs are going to stop soft-pedaling the anti-NATing provisions in their terms of service
    Check SpeakEasy DSL. Not only don't they care about NAT, they don't mind wireless connecting sharing either, so long as none of the people (ab)using your connection are violating the AUP. As with any ISP you're responsible for anyone using your connection. In other words, setup your WAP, secure it to trusted friends and family only, and you're A-OK.

    And I'm not even a SpeakEasy customer. I wish RoadRunner would implement similarly "with-it" policies...

    • Agreed, SpeakEasy ROCKS (Score:5, Informative)

      by phillymjs ( 234426 ) <slashdot.stango@org> on Sunday December 22, 2002 @06:26PM (#4942024) Homepage Journal
      I dumped my Comcast cable modem for SpeakEasy DSL just about a year ago, and I've never been happier with an ISP.

      I've never had an outage, I can (and do) run servers in my house to handle my own mail, www, etc. Their phone reps are courteous and don't suffer total brain shutdown when I mention the magic sentence, "I'm using a Mac."-- not that I've needed support beyond a little help sorting out a password issue when I wanted to configure the INCLUDED dial-up access in case I needed it.

      When I needed an additional IP address last year, I called them up and had one in a couple minutes. I just upgraded my service from 608K down/128K up to 1.5M down/384K up, and it was not a hassle at all. Since my employer chips in $50/month of my ISP charges, this higher speed service now costs me $40 per month, the same I was paying a year ago for my cable modem with all of its restrictions.

      To all Slashgeeks who are able: If you have a choice, go with SpeakEasy-- you won't regret it!

      ~Philly
      • not all of us can afford the high costs of DSL. I personally have never been happier than I have been w/cable (ATTBI sucks only a little, RR was great).

        768k/128k is entirely too slow and too expensive. Not all of us have the $50 chip in offer.

        I run servers, I have mutliple NAT'd computers, and I use an AP for my laptops.

        No complaints from ATTBI. Just don't do anything stupid (like hack your modem cfg file) and you are fine.

      • I've posted my woes about Speakeasy before, so I won't beat a dead horse. However, just a quick recap -- nightly outtages, billing snafus, unhelpful customer service, malicious tech support (waiting until my modem was out of warranty to diagnose my nightly outtages as a modem problem, even though the modem did that from day one), and more. Sure, I can and do run my own servers, and the tech support guys don't shit a brick when I tell them how I have my network setup (it's convoluted, involving a software bridge, a switch, a hub (I'd prefer a switch here, but I had the hub on hand), an AP, and quite a few computers and appliances). They also give me a lot for my money, and I'd be hard-pressed to find a DSL service that provides me exactly what I need as Speakeasy does. However, they're far from perfect.


        To all Slashgeeks who are able: If you have a choice, go with SpeakEasy -- you might regret it, but you'll never know unless you try, right?


        I guess that wasn't so "anti-Speakeasy" after all. Moral of the story: "Buyer beware."

    • I wonder when ISPs are going to stop soft-pedaling the anti-NATing provisions in their terms of service

      When people stop paying them to do so and start paying ISPs who don't get a kick out of deceiving and walking all over their customers, as AT&T and RCN have done to me. I can't even begin to express the comfort it has given me to switch to Speakeasy.

      It's not that they don't care about wireless access. They actually encourage it. Imagine that, a company that wants you to get the most out of the service you're paying them for, unlike most of the rest of the industry, which wants you to use your connection as little as possible, so that they don't have to spend any of the money you're giving them on their network, and browbeats you when you try to do anything more than passively read web pages and email.

    • As someone who has worked for an ISP, I can tell you that the majority of the traffic comes from a minority of the customers. I haven't worked for an ISP that's had DSL service, and so, I see it from the point of view of dial up. In fact, I'm going to assume that the AUPs in question stem from dialup related problems.

      You see, the issue is that phone lines cost money. For us, it was costing us almost $70/month, [business lines are more expensive than residential lines, and well, they were PRIs], plus the cost of hardware (modems, servers) and other costs (staffing, office space, internet connection). The only way to make a profit off of modems is to NOT have someone sitting on 'em 24hrs a day.

      Although we didn't get picky when a whole family was using their connection, we would start asking questions when someone was on more than 500hrs in a month (as well, we expect people to sleep once in a while). Sometimes, you get more than one person in the house who's an addict, and they're working different shifts... normally, it's someone who has their mail program set to check every 5 minutes, and they've got dial-on-demand set up.

      So, to make up for this, and so you didn't have to completely kick people out, there were normally additional tiers of service. Many times, they were just listed as 'business' and 'residential', or they might have some other name for an always-on-connection, or where it was okay to be sharing out the line -- because the more computers using the line, the more likely it was to be up all the time.

      Now, with DSL, the model has completely changed, so that doesn't entirely fit, but have the ISPs changed their AUP? Doubtful. And if they did, they'd probably have to have one for broadband connections, and one for dialup, and then you've got to have seperate tracking of users, etc. Of course, your bottleneck problem now isn't busy modems, but your connection to the internet -- so people downloading every mp3 they can find, divx movies, etc, start cutting into your bottom line. [And well, you've either got to lose money on each individual that's doing it, or get rid of 'em so they don't adversely affect every other customer.]

      Oh, and for those wondering...I am a speakeasy customer...only outage I've had was when they changed my IP address. [and they even called me the day before to remind me it was going to be happening... but I wasn't at home, and forgot to write down my new IP, and I had it saved in e-mail, so I had to go to the local library the next day, and check my mail from there to get my new IP]

  • NAT is hard to detect (Score:5, Insightful)

    by Gothmolly ( 148874 ) on Sunday December 22, 2002 @06:12PM (#4941977)
    Unless you start snooping into the data packets, looking for User-Agent strings, etc (if you're NATting 2 different OSes behind it). You can make educated guesses (why do all TCP connects come from port 61000+?) but you really can't KNOW if someone is NATting.
    • This is somewhat easy to find people that are extremely likely to be NAT'ing actually, usually with no extra hardware than what the ISP already has.

      You can look at the source port of the packet (going from the customer to wherever, i.e. ingress to the ISP's network), to be fairly confident that the customer is NAT'ing their traffic.

      Most NAT implimentations change the source port to a very high port number (usually in the 60,000+ range) when they translate (along with changing other info). You can probably be fairly safe to log anything above 40k (although some research into what more popular 'ethernet routers' that do NAT use for port ranges would be prudent). Normal traffic for most implimentations of a TCP/IP stack won't typically source off of ports that are reserved for NAT'ing, so again, you can be fairly sure they are NAT'ing if you 'catch' them (enough to call them to ask them about it anyway.. that way if they aren't, you don't shut down a paying customer for no reason).

      On a cisco, you can setup an extended access list applied to the interface the traffic is hitting, with logging enabled to see who is NAT'ing, see if it continues for a while, and if it does, you can be fairly assured they are NAT'ing traffic.

      Mind you, there are many ways around this for some implimentations of NAT (i.e. changing the port range NAT'd packets source off of), but for average Joe User hooking up an 'ethernet router' that does NAT, they could be caught fairly easily by this (and other methods that work in conjunction with this).

      (this is just a brief summary, don't blame me if it isn't detailed enough).

      • Re:NAT is hard to detect (Score:1, Insightful)

        by Anonymous Coward
        It is just an extended writeup of what the original poster already said. It is a method that is not guaranteed to work, and as soon as it is used you will see new router firmware that circumvents it.
        (there really is no "port range reserved for NAT" and it isn't required either)

      • Not true any more (Score:2, Interesting)

        by wotevah ( 620758 )
        The new NAT mechanism implemented in the Linux 2.4 kernels tries to use the same source port as the translated packet. I suspect other NAT implementations might be doing the same.
    • Would running a proxy server be harder to detect? I run squid and have all of my home pcs connect to the web through it.

  • That's a nice-looking amp... (Score:5, Funny)

    by BitHive ( 578094 ) on Sunday December 22, 2002 @06:16PM (#4941989) Homepage
    ...but ours goes up to 11.

  • God forbid. (Score:5, Insightful)

    by mwillems ( 266506 ) on Sunday December 22, 2002 @06:17PM (#4941994) Homepage
    The entire point of 802.11b is that it is low power, allowing a lot of use in a small area. If my neigbours all start to boost their signals (both ways of course), my 2.4 GHz RF environment is going to be be useless. X.10, cordless phones, bluetooth, 802.11b all share this frequency (and I have all four of those at home...)

    In the UK, until recently 802.11b was illegal for commercial use because it was full, making it useless. My car the other day would not open in a high-RF environment (near Toronto's CN tower).

    Can we please keep this frequency useable? Amplifying will kill it. An RF signal of this type can carry 30 miles very easily, making it useless if we all start amplifying.

    Michael (radio amateur, VA3MVW)

    • URL to his obit. page (Score:3, Interesting)

      by tytso ( 63275 )
      I don't think you need to worry. What Linksys is doing is not nearly as interesting as people might assume (which is I suppose is par for the course for Slashdot :-) What I'm pretty sure is going on is that the wireless access point on the linksys doesn't have a very strong radio transmitter to begin with (I'm guessing 30-50mW), and the signal amplifer just raises the transmit power to the max legal limit for the 2.4GHz band.

      The Cisco 350 Access Point (and wireless cards) has better receive sensitivity (I don't know if that's due to a better built-in antenna, or better radio circuity, or both), and a stronger transmit power than most other 802.11 cards (selectable from 5mw to 100mW). In contrast, the Lucent Wavelan Silver card has a 31 mW transmitter. I don't know what the transmit power for the Linksys access point, since it's not listed on the web site or in the user's guide, but they claim an outdoor range of 1500 feet at 1 Mbps, and 500 feet at 11 Mbps. For comparison, the Orinocco access point claims 1750 feet at 1 Mbps, and 525 feet at 11 Mbps, and the Cisco 350 access point claims an outdoor range of 2000 feet at 1 Mbps and 800 feet at 11 Mbps. If we assume that both Cisco and Linksys are exagerating to an equal extent for the best case scenario, it seems pretty clear that the Lucent transmitter is less powerful than the Cisco 350.

      Of course, as radio amateurs know, transmitter power doesn't have as much effect on range as some people might think. That's why QRP operators can sometimes communicate with people halfway across the globe with only a Watt or two of power. So the Linksys signal amplifier will probably not make that much of a difference.

      That being said, I would recommend the Cisco 350, not because of the higher transmit power, but because the access point has better manageability (you have much finer control over how the access point operates, with various nice features such as having the AP ask your radius server whether or not a particular MAC address should be allowed, LEAP authentication/encryption, etc.). Also the Cisco 350 PC card has a full-featured Linux driver, which allows you to control the transmit power, scan for all available 802.11 networks, and so on. Another nice feature with the Cisco 350 is that you can store the WEP keys in flash memory, so that you can lend the card to house guests, without needing to reveal the WEP key. (Right now, I haven't been able to find an open source radius server that supports LEAP, so I'm using a combination of 128-bit WEP keys plus MAC address access controls. One nice thing about the 350 Access Point, as compared to the Apple airport, is that you can change WEP keys without needing to reboot the access point. So while I haven't implemented it yet, it should be possible for me to automate changing the WEP key every 24 hours, by calculating a MD5 hash of a secret plus a timestamp. That way, a shell script on my Linux laptop would allow me to get update the WEP key at the same time, automatically.)

      -Ted (N1ZSU)
      • Ted,

        Mmm.. I wish I could be as much of an optimist. I have recently noticed that 2.4 GHz seems to be getting fuller. My MAME box in the basement sometimes will not see the wireless access point upstairs, and I suspect it's due to QRM - in other words, to the neighbours doing stuff (X.10, phone, etc) in the same frequency range. Power isn't everything (with 10W on 6m I have worked SCotland), but this is line of sight, and surely raising the power from 40 mW to 250 mW will enable more distant neighbours to add to that problem. Bluetooth uses 1 mW by the way, which gives 10m range. Power isn;t "nothing" either!

        73
        Michael

    • Re:God forbid. (Score:5, Funny)

      by phunhippy ( 86447 ) <zavoidNO@SPAMgmail.com> on Sunday December 22, 2002 @07:20PM (#4942183) Journal
      My car the other day would not open in a high-RF environment

      did ya consider trying the Key?? its a nice shiny metal thing they give you with your new car :)

      • Re:God forbid. (Score:3, Insightful)

        by frohike ( 32045 )
        Yes, that's funny :), but a lot of cars these days come with a rudimentary alarm that will get irritated at you if you try to open the doors without having used the remote, and you locked it with the remote. It's a really simplistic defense against slim jims and the like.

      • Re:God forbid. (Score:3, Insightful)

        by shadowj ( 534439 )
        Mercedes models don't come with the usual shiny, metal things. The "key" is actually a vaguely key-shaped RF and IR transmitter. A small traditional metal key is buried inside it, and can be used to unlock the door in case of a failure... you have to pry a plastic cap off the hidden keyhole on the door. The emergency key won't start the car.

    • Re:God forbid. (Score:3, Informative)

      by Lumpy ( 12016 )
      RF signal of this type can carry 30 miles very easily, until it hits something that has a strong adsorbtion rate at 2.4Ghz something very uncommon and non-abundant like..... water...

      you haven messed very much with wifi or anything in the microwave range have you.

      I have a 1 watt transmit amp on each end of a 23 DBi gain dish link... IT WILL NOT FIRE THROUGH A TREE because of the adsorption rate of water. I wished I could burn through a tree but I cant not even with the huge power (as far as wifi is concerned) I am using... (shhhh dont tell the FCC I'm violated every fricking law with this expieriement)

      it will NOT travel 30 miles, it will NOT kill everything or cause huge problems for anything but low end gear that barely works anyways... (your car keyfob/car alarm opener for example.. it dont matter who makes it they are low grade junk in regards to the RF functions of it.)

      I can show you at least 5 houses that have a field strength at least 10 times that of anything you could buy in 802.11b equipment... and it's because of MICROWAVE OVENS that are old and leaky..

      please dont fearmonger... these tiny toys that linksys are selling wont hurt anything... and 95% of those out ther eamplifying any of their 802.11 gear are not capable of doing it anywhere remotely correct to get anything but a marginal gain.. Hell some of these people are using the worse feedline and antenna designs that are attenuating more than they gain...

      stop worrying... the no-code ham licenses has more of a chance at ruining the airwaves than amplifying 802.11 traffic.

      • Actually, I have played with microwave up there a bit, yes, and I do have WiFI all through the house; and although this absorbs a little more than, say, 440 MHz (which I use all the time), it certainly goes through "things" and trees. Yes, it is attenuated, but it does go through. And as for not travelling 30 miles: it will travel 3000 miles, or 30,000 for that matter, with the right antenna and line-of-sight. Ask NASA.

        As for the FCC, well it (and its equivalent in other countries) has all those laws for good reasons. They're there to make comms possible, not just to hassle us for no reason.

        Simple fact: my neighbour's X.10 cam and cordless phone interferes with my use of the frequency, and the higher the power, the more neighbours get to interfere.

        Mike VA3MVW

      • Just read a web site about a ham "working 500kms on 3cms with 2.5W on Rain Scatter". That is 500 kms' not the 50 km I was talking about. Rain scatter!!

        See http://www.neoamateur.org/

        Mike va3mvw
        • a) 2.5W = 2.5 times the FCC legal limit for 802.11

          Also, thy were probably using a very high-gain antenna on top of that 2.5W of transmit.

          Last but not least - They were almost surely using CW (or *MAYBE* PSK31) - Which have bandwidths measured in *hertz* not megahertz.

          The closest comparison to this would be taking a magnetron from a microwave, putting it into a feedhorn, and using that to communicate.

          (Actually, hams DO things like this - Use a PLL to clean up the magnetron's signal, and then use it with a big dish for EME (Earth-Moon-Earth) work)
    • He bought one of the X10 cameras from the evil advert! It is the only product I know of from X10 that uses the 2.4 Ghz Band. The wireless RF light controllers are below 400 Meg. I forgot the exact spec, but I believe they are in the 300 Meg band. The carrier current remotes are near 120 Khz.

  • Amps? Erk (Score:2, Insightful)

    by FrostedWheat ( 172733 )
    I don't like the idea of amps for 802.11.

    People seem to be getting excellent range out of very low power devices using fairly cheap antennas!

    Pumping out more power will just increase interference with other 802.11 networks in the area. Not to mention it would almost certainly be illegal in the UK to use one of these things :)

    • Re:Amps? Erk (Score:2, Insightful)

      by chriton ( 29476 )
      I have to agree. I have two high gain directional atennae; one at each end of my wireless link & I get 22Mbps all the time (it's a DLINK using PBCC). Before the antennae, the link was dodgey at best. Now, without increasing the power I'm putting out to the neighborhood, I have exactly what I need. (Well, I suppose some neighors in some particular directions are getting a little more, but some less as well.)

      Amplifiers just make things more & more crowded. I have too much stuff running in the 2.4GHz range already and do occasionally have some interference problems, especially with the cheaper of my cordless phones.

      I know it's legel, I just don't like it much.

  • FCC regulations? (Score:1, Interesting)

    by Anonymous Coward
    What are the FCC regulations regarding the maximum legal RF Signal strength allowed in these wireless devices? Aren't they under the part 15 regulations (low power transmissions).

    Or are the regulations written to put a cap on the intensity or "Field strength" of the radiations like the limits in the FM band (88-108mhz)?

    Can someone comment on this?

  • Is a signal strength war already escalating? (Score:3, Interesting)

    by phr2 ( 545169 ) on Sunday December 22, 2002 @06:29PM (#4942031)
    A few people are always trying maximize the range of their WLAN's, reach through more walls, etc. But are there enough to support a "mass market" product like this? Or does it mean people are looking for more powerful AP transmitters because they're trying to overcome interference from other WLAN's in the same office building? And will those other WLAN users jack up their own transmitters in response? I wonder if we're seeing the beginning of an RF power output war, sort of like we already have between cell phone carriers with towers along highway 101 (in Calif.) and similar freeways. This is NOT good. We get cooked by enough microwave emissions in our offices already.

    I'd be a lot happier if the FCC got rid of some of those UHF TV channel frequency monopolies and gave the spectrum back to the public that rightfully owns it, to try to lower contention over the narrow strip of bandwidth that 802.11 uses. There's also much more use of wireless these days by non-mobile devices than there really needs to be, when those devices could perfectly well use wired ethernet, or maybe a much lower-powered shorter-range wireless scheme (like a higher-bandwidth Bluetooth) to an access point that's nearby (i.e. in the same room or close to it, not far away in the building). However, that last part is harder. Tragedy of the commons and all that.

    • Since 11Mbps is a tasty target, it's hardly surprising that people will be gunning for maxing out what 802.11 will do.

      Maybe the better thing to have done, rather than come out with a single wireless protocol would be to have two; one that would do 50Mbps but at a range of no more than 50m, and another good for 50 miles but at no more than 250Kbps, with a large number of channels.

      The former would be great for offices or other places that need high bandwidth, but the distance limitation would have kept it from being so popular as a last-mile. The latter would be awesome for linking buildings or other long-range applications, but a large channel count and low bandwidth would keep people from trying to replace T1s between buildings..
  • The Wap11 Hack [seattlewireless.net] courtesy seattle wireless.

    Basically using a different firmware/snmp agent for a different ap based on the same chipset you can up the signal strength at no charge. It is risky however, so I take no responsibility blah blah blah. Maybe thats all the amp does is up the built in power.. ;)
    • Only works for the WAP11 v1.x. Won't work on the WAP11 v2.x or any of the combo router units. I've got a WAP11 v2.2, and I'd be delighted to use my wireless connection in my living room...even 50' away, the connection is sunk.

      I suspect the amplifier isn't limited to LinkSys products either, although its design is meant to fit with them.

  • more info (Score:5, Informative)

    by trb ( 8509 ) on Sunday December 22, 2002 @06:53PM (#4942108)
    That link didn't show much (at least for me). You might want to try WSB24 [linksys.com], which points to a data sheet. [linksys.com]
    • Peak Gain +14 dBm
    • Receiver Gain +20 dBm (nominal)
    • Tx Input Power +5 dBm (minimum) - +20 dBm (maximum)

  • big deal. ydi.com and rflinx.com have quality amps (Score:5, Informative)

    by puzzled ( 12525 ) on Sunday December 22, 2002 @06:53PM (#4942112) Journal

    A while ago on slashdot there was breathless coverage of a Linksys hardware 'hack' that raised their 30mw AP to 100mw. Some knowledgeable RF engineer took a look and it *was* putting out 100mw - 31mw in channel, and 69 mw of crap spattered all over 2350 - 2550 MHz. All that 'hack' accomplished was giving more ammunition for those satellite radio folks that want to regulate the ISM band.

    Instead of building a 100mw radio with good sensitivity, Linksys is building a cheesy amp to go with their cheesy AP.

    If you genuinely need some amplification I've used YDI.com and Teletronics.com amps in the 250mw to 1 watt range and not had much trouble with any of them. RFLinx or RFLynx(sp?) has come out with a 750mw amp for $200, but I haven't tried that product yet.

    FYI half of the reason to deploy an amp is for the LNA (low noise amplifier) effect - besides boosting output they pump up the return signal by 10 - 14 dB. There is a real call for a 150mw output amp with a solid LNA for client side problems, but that is a story for another day.
  • I wonder when ISPs are going to stop soft-pedaling the anti-NATing provisions in their terms of service.


    I wonder when ISP's are going to realize that it is futile to say "no NAT" or "no servers" or "max transfer gb per month" and realize that the only sane thing to do is to provide unrestricted access, and simply charge their customers what it actually costs to provide xx mb of bandwidth?

    Let's not beat around the bush. Heavy users pay for heavy bandwidth. Light users pay for less bandwidth, and get less bandwidth. Trying to weasel out of providing less than the amount of data that the pipe can carry is a waste of everyone's time.
    • I wonder when ISP's are going to realize that it is futile to say "no NAT" or "no servers" or "max transfer gb per month" and realize that the only sane thing to do is to provide unrestricted access, and simply charge their customers what it actually costs to provide xx mb of bandwidth?

      Because, if Company A has "UNLIMITED" in really big letters, but all of the restrictions (no servers, etc) are less prominent, and Company B says "We allow you to have servers, but we charge money per mbit" Joe user will go with company A almost everytime. Why?

      Because "UNLIMITED" stands out, unlimited is certainly better than limited right? The average broadband internet user doesn't have a server, and chances are, they don't know how many mbps of traffic they make use of in a month. Joe user will see the part from Company A, not care about the restrictions (the fact remains that the majority of cable modem/dsl subscribers /don't/ have servers), and choose it over Company B because "UNLIMITED" means better to most people.

      The 'better' answer is to have multiple tiers of service, where the base plan is unlimited, but with restrictions (no servers, one computer, etc), and another plan where you can have servers, but only if you pay extra (per mbps over a certain amount, etc)... The problem is, most people who have a '*NIX box' with a web server wont want to pay extra, and will just get the cheaper plan, so 'policing' (port scanning their address space for people running servers on well known port numbers) for people running servers, would be necessary. Also, it would probably be prudent to have a 'power users' plan, where if you have over X computers (NAT'd or not) you have to pay X amount. Mind you, this would suck for a lot of people, since they'd have to pay more per month, but overall the cost of broadband for the average user should go down. It makes sense, it just depends on how you market it.
  • That amplifiers amplify everything... including the ambient noise PLUS they also insert a certain amount of noise themselves (LNA stands for "low-noise-amplifier not no-noise-amplifier). So while the signals may be amplified, the noise level might also be amplified enough to negate the effect.

    In addition, the FCC has a dba limit on the amount of signal you can have so you cannot just stick power amplifiers on all the devices in the network without incurring some exposure to fines and penalties (for interference).

    It makes a lot more sense to design the wireless system to use numerous low-power devices spread around the area so that you can cover just what you need to cover and not simply saturate the area with signal.
  • Say all you wise wi-fi-ers, can someone help me with some advice? My sister has an 802.11b router at one end of her house and a desktop with a wireless card at the other end (it's a long, skinny ranch style). She couldn't get any signal on the computer with the card. She went out and got an access point to set up in the middle of the house, but had trouble getting it to work. She finally got the two boxes talking by putting the AP at the end of 50 ft of cat-5... just inside the room with the desktop. Kinda defeats the purpose of having a wireless setup. She's been planning on figuring out how to make it all work better, but is probably still stepping over the cable on the floor everyday. So, how can she boost the signal? Will an AP between the two work as a repeater? What's the simplest and cheapest way to get this to work? Any sites I can check out that'll help answer these questions (without making my head spin)?
  • Well from what i understand Comcast does try to inforce the provison on open access points and multiple computers w/o paying.

    My friends on Time Warnmer on the other hand told me that TW installers tell 'em taht a router is the way to go. TimeWarner also doesn't block port 80 or scan it like comcast seems to be doing to me, same subnet as Comcast Name server is how i determined it was them and not some jerk trying to break in(i know you are thinking that). People on VerizonDSL also tell me that Verizon installers also don't care and even recommed using the routers due to the natural firewall.

    I hyope ISPs realize charging by the computers is stupid. I mean like when i get an OC3, what a dream come true, they wouldn't charge me for every computer on that.

    Also Comcast is dumb, horrible customer service & poor product service and pricing, if DSL becomes available for me as Verizon tells me it will. I'm ditchin comcast for satelite and VerizonDSL.

    Have a good New Year All. Comcast you may want to let your customer service reps know taht you are performing upgrades instead of having the guy trounleshoot with me on the phone...exchange modems...send some guy out after sending one to my neigh boor...to tell me they are upgrading and repairing some damage due to heavy rains.
  • No, really, why?

    Why does my ISP care how many computers I hook up to my network? I'm paying for the bandwidth.

    If I pay for the bandwidth, it doesn't matter if I'm using 1 computer or 20, the results are the same.

    The best thing for an ISP to do is to charge customers for their actual use, just like the electric company. None of these silly no servers, no sharing, etc. clauses.

    Would you like it if the phone company said you could only have 1 phone on your line, or the power company said you can't share electricity with your roommates? I didn't think so.
    • > Would you like it if the phone company said you
      > could only have 1 phone on your line,

      The phone company did just that for more than half a century. In fact, they did not allow customers to use their own phones at all. We were supposed to rent phones from them.

      I agree, though. ISPs should sell metered service and charge by the byte.

    • Why does my ISP care how many computers I hook up to my network? I'm paying for the bandwidth.

      No, you're paying for a personal internet connection. If you were paying for bandwidth, it would likely be more expensive and they wouldn't care what you did with the connection. That said, it shouldn't matter what you do, so long as you don't abuse the service, say by downloading stuff 24/7 or having 4 or 5 people using it all the time.

      Would you like it if the phone company said you could only have 1 phone on your line, or the power company said you can't share electricity with your roommates? I didn't think so.

      Well, it's already been mentioned that the phone company did just what you describe. Your analogy is flawed. Having more than 1 phone on the line doesn't affect the load on the phone company - you can only make 1 phone call at a time. The power company charges you based on usage, and at a higher rate if you exceed a set amount per mnonth, so why should they care?

      • > No, you're paying for a personal internet
        > connection.

        But what I want to pay for is bandwidth usage.

        > If you were paying for bandwidth, it would
        > likely be more expensive...

        Why?

        > ...and they wouldn't care what you did with the
        > connection.

        That's the point.

        > ...it shouldn't matter what you do, so long as
        > you don't abuse the service, say by downloading
        > stuff 24/7...

        Not likely with one phone line used for personal calls, business, and the computer.

        > ...or having 4 or 5 people using it all the
        > time.

        Two people, and usage that is probably below average.

        • Buy what I want to pay for is bandwidth usage.

          That is not what you bought. Your options are: business class DSL, various kinds of leased lines, and colocation. Wireless is sometimes an option, but that isn't widely available.

          Why?

          All your options for paying based on usage are expensive. If you just want a server, colocation and managed hosting are likely your best bet.

          Two people, and usage that is probably below average.

          So why are you bitching? Your ISP may care, but it can't find out and, unless you make a big stink, they won't care about you beyond your monthly check. In all seriousness, if you care about your ISP and its posturing, then switch. If you can't, then keep your head down and donate to a lobbyist group that represents your interest. That's how Democracy works in the US.

  • Wow, what crap. (Score:3, Interesting)

    by Wakko Warner ( 324 ) on Sunday December 22, 2002 @07:48PM (#4942278) Homepage Journal
    I bought a 12dB omnidirectional wireless antenna for like $100. This thing doesn't even give specs (that I can see) -- and I'm willing to bet it costs more than I paid for mine. Linksys, as usual, sucks.

    - A.P.

  • Linksys (Score:2, Informative)

    by wakeboard ( 556264 )
    What I wish linksys would allow in the firmware was a mac address allow list. Right now you can only set mac addresses to block from the WAP. Kinda stupid if u ask me, why not list the mac addresses you want to allow. I know that they can be spoofed, but it makes it just THAT much harder
  • Most folks I know of here in Columbus who have more than one computer use NAT. Mostly Roadrunner. A tech was out and seen my setup and they said nothing. Only thing they said was t would be slower (BS). In facet I think they support you doing things like NAT. Saves them equipment and IP adresses.
  • YMMV, but SBC Yahoo / Ameritech doesn't care about NAT. In fact they acknowledge that customers may be connecting more than one system in their customer support materials. The downsides:

    -Non windows / mac OS support doesn't exist. They run a don't ask don't tell policy when it comes to multiple PC's, but you need a windows PC for them to do any sort of trouble shooting. (My guess says their support people are reading from a card)

    -Installation (from the telco side) is intermittent. Some people I've talked to get setup in a couple weeks. My setup took just shy of three months. To be fair, my order went in just a few days after my area went DSL ready. half of their systems said I could have DSL, the other half didn't. I had two separate orders canceled by the compliance check.

    SBC Yahoo may be a good option for your area. sure comparing it to Comcast / Roadrunner is like comparing genital warts to Leporasy (you don't really *want* either, but one's probably a bit easier to live with). It may be wrth investigating. I'm running wireless + two regular boxen and they (officially no less) don't care.

    And if you do setup a network remember; http://www.coyotelinux.com, because if you buy a router from best buy, then the terrorists have already won.
  • I wonder whether Linksys will include an upgraded antenna in their signal booster kit instead of the little rabbit ears that their current access points come with. For about $100 you can get a 10 decibel gain antenna (for example, see www.telexwireless.com.) Add a 10db gain omni to an amped signal and we may really be able to get some distance out of 802.11b!

    What the world needs is fewer karma whores and more good friends.
    Go ahead, friend. :) Click that white button and turn it green.

Slashdot Top Deals

Solutions are obvious if one only has the optical power to observe them over the horizon. -- K.A. Arsdall

Close