First, let me explain that the wireless LAN technology that I am referring to is radio based, not Infra-Red (IR) or Microwave. The IR version requires “line of sight”, as does the microwave version, which I consider to be a major drawback.
It can, however, support faster throughput. The microwave specification is aimed more towards ISPs (or possibly large corporations that require *LOTS* of bandwidth) than your typical network.
The specification (for those of you who like white papers on protocols) for the radio version is IEEE 802.11. This specification, depending on the frequency used, allows for wireless network bandwidth of up to 11Mbps (remember that 1Mbps = 1024Kbps).
The most widely used technology for (radio based) wireless LANs is known as Spread Spectrum. The US military invented this technique to avoid jamming and eavesdropping. That is good to know because network security is still a concern; a hacker now has to intercept radio transmissions on several frequencies before he/she can even reassemble the packets he/she is trying to intercept.
There are two basic types of Spread Spectrum technology being used: “frequency hopping spread spectrum (FHSS)” and “direct sequence spread spectrum (DSSS)”. The distinctions are quite technical at this level, so if you would like to read more check out The Wireless LANs Page. This document is slightly dated but gives a good feel for the wireless technologies and how they work.
You are saying, “It has to have a catch,” aren’t you? There are a couple of small catches. The main hangup is that, like any other type of radio, interference is a factor of range. Normal conditions will provide you with a range of up to 100 meters indoors and 300 meters outdoors. Normal 10base-T ethernet requires a repeater every 100 meters, so this probably will not affect you any more than your typical wired network would. Again, like any other type of radio signal, steel buildings or locations near power lines may not be suitable for transmission; at least not at the maximum advertised range. You will have to check.
The next thing that could be a problem for some of you is the bandwidth available. Most networks do not need bandwidth in excess of 11Mbps; in fact, until recently (prior to 100Mbps Fast Ethernet), ethernet operated at 10Mbps and the fastest token ring has ever operated at is 16 Mbps. In most cases you are not going to have Internet access in excess of 11Mbps. Unless you are moving many large files across your intranet or have many machines on each segment of your network, you probably do not need much more than 11Mbps.
Great, so what do you need to make it work? You will need an access point. This piece of hardware acts as a bridge and a hub combined. On one side it will transmit and recieve data to and from your wireless nodes. On the other side it has your garden variety 100Mbps Fast Ethernet for connection to an ethernet network, or, via a crossover cable, to a single machine (server, I’m guessing).
For those of you wondering about DHCP and BootP capabilities, these devices are usually BootP enabled, so no special configuration for DHCP and BootP clients will be necessary. I have seen 11Mbps Access Points for as little as $200 here: www.pricewatch.com(networking, network cards, wireless).
Next, you will need your client devices. These are available as either PCMCIA or PCI adapters. These will communicate with the Access Point(s) to provide your network connectivity. You can find 11Mbps wireless network interface cards (WNIC ?) for as little as $95 here: www.pricewatch.com (networking, network cards, wireless).
Another thing that is really cool (and borrowed from the cellular phone industry) is the idea of cells. A cell is the area around an access point that will allow communication with wireless clients. (ie. Picture the cell as a sphere with the access point at the center. The surface of the sphere would be the same distance away from the center of the sphere as the effective range of the Access Point.) If you were to overlap cells, you would be able to “roam” between cells with no loss of transmission. Pretty cool, huh?
On a final note, I cannot find much documentation on compatability amongst manufacturers of these devices. With that in mind, I believe it would be wise to use components from the same manufacturer in order to avoid compatibility issues. I am sure that when this technology is not so new anymore that this will become less of an issue.
Speaking of issues, that is the end of this one. I hope it has been useful to you. Remember, any questions/comments/suggestions are always welcome.
Jay Fougere is the IT manager for the Murdok network. He also writes occasional articles. If you have any IT questions, please direct them to Jay@https://www.murdok.org.