In my November 2003 column, I reported to you that the ISO/IEC TR 24704 Information Technology Customer Premises Cabling for Wireless Access Points was a work in progress. ISO/IEC TR 24704 was written because the infrastructure specified in ISO/IEC 11801 Ed. 2.0 did not specifically address connections to wireless access points (APs).
ISO/IEC 24704-2004, published in March 2004, extends the ISO/IEC definition of premises cabling to support a wide range of wireless technologies, including the Institute of Electrical and Electronics Engineers (IEEE) 802.11 wireless local area networks (LANs), Digital Enhanced (former European) Cordless Telecommunications (DECT), and Bluetooth personal area networks (PANs).
Know your alphabet
The IEEE 802.11 standard specifically addresses wireless LANs. The “a,” “b,” and “g” notations identify different variations of the IEEE 802.11 standard. IEEE 802.11b, the first version to reach the marketplace and the slowest and least-expensive of the three, transmits at 2.4 GHz and can handle up to 11 Mbits/sec. Next was IEEE 802.11a, which transmits at 5 GHz and can handle up to 54 Mbits/sec. And the third was IEEE 802.11g, sort of a mix of both worlds, transmitting at 2.4 GHz (giving it the cost advantage of 802.11b) and also having the 54-Mbit/sec speed of IEEE 802.11a.
ISO/IEC 24704-2004 concentrates on the planning considerations necessary for future connections to wireless APs that supplement, not replace, the existing copper and optical-fiber premises cabling system.
ISO/IEC 24704-2004 uses a honeycomb wireless cell structure-it looks like someone overlaid the building plans with a bee’s honeycomb-and recommends a cabling grid spacing of 12-meter radius to accommodate the operating range of some radio technologies.
A minimum of one Class D (Category 5e) telecommunications outlet is recommended per coverage area (cell). According to the TR, the total number of outlets for the building should not increase by more than five percent … read that as plus-five percent on the cabling budget.
Yes, these wireless documents are about more cabling in a building, not less.
ISO/IEC 24704-2004 also describes how the wireless APs can be either remotely powered over copper cabling from a telecommunications room or midspan-powered as long as the power supply is external to the permanent link. This is further described in IEEE 802.3af DTE Power via MDI. If optical fiber is used, local powering would be required, or you could install an unshielded twisted-pair cable for power.
Because predicting microwave propagation within buildings can be more art than science, especially at the design-documents stage, ISO/IEC 24704-2004 recommends the use of radio frequency site surveys in addition to the cabling grid. Translation: design the cabling infrastructure per the guidelines and then test to be certain that every space is covered.
Of course, that would indicate that you would actually have to connect wireless APs to those cables to see if the cells were covered. Hmmm… that is a lot of testing that needs to be accounted for in your labor contract. And if you are using one or two wireless APs for the testing, do not forget your power drop cords, as they will not likely be line-powered at this stage.
ISO/IEC 24704-2004 does not address the placement and security of wireless APs, only the cabling installed to link and power them.
TIA’s TSB-162
And now, TIA TSB-162, which addresses the very same issues (but not necessarily in the exact same way), has also been released for publication.
TIA TSB-162, Cabling for Wireless Access Points, provides guidelines on the topology, design, installation, and testing of telecommunications cabling infrastructure, in compliance with ANSI/TIA/EIA-568-B.1 and TIA-569-B, for supporting wireless LANs in customer-owned premises. TSB-162 covers the cabling between LAN equipment and wireless APs, including pathways and spaces to support the cabling and wireless APs.
The primary difference is the references to TIA documents, and the cell is a square rather than a hexagon. While the change appears to be subtle, it increases the number of cables that will be required to cover a building floor plan … read that as plus-five percent; plus, on the cabling budget.
And so this begs the question, “Why do we need two documents addressing the very same issue?”
The majority of the folks who participate in the writing of these documents are engineers. And it has been my personal observation over the last 35-plus years that, to an engineer, all matter in the universe can be placed into one of two categories: Things that need to be fixed, and things that will need to be fixed after you’ve had a few minutes to play with them.
You see, engineers like to solve problems. And if there are no problems handy, they will create their own problems. I am sure that you have heard the saying, “If it ain’t broke, don’t fix it.” Well, engineers believe, “If it ain’t broke, it doesn’t have enough features yet.” To an engineer, the world is a toy box full of sub-optimized and feature-poor toys, each of which needs at least a few good technical documents written so as to confuse the user about which to follow.
Both documents are available for purchase from Global Engineering Documents (global.his.com).
How do they do that?
With all the talk about WiFi these days, have you ever wondered how it actually works?
Do you remember playing with walkie-talkies when you were a child? Walkie-talkies are basically small radios capable of transmitting and receiving radio signals. When you speak into a walkie-talkie, your voice is picked up by a microphone, encoded onto a radio frequency, and transmitted out the antenna through the air where it is received by the antenna of another walkie-talkie, then decoded from radio signal back into voice, which can be heard via the speaker. Walkie-talkies normally transmit at 49 MHz, at a signal strength of about 0.25 watts, for between 500 and 1,000 feet.
Now, picture connecting two computers together in a network using this technology; each computer would have a “walkie-talkie” type device that can be switched from transmit to receive. Only instead of voice-into-radio signals-back-into-voice, we would encode binary 1’s and 0’s into radio, then back into 1’s and 0’s. This would actually work but would be incredibly slow-maybe 1,000 bits/second or so, but you get the idea.
The same, only different
Like the radios used in inexpensive walkie-talkies, radios used in WiFi’s radio technology have the ability to transmit and receive, and they have the ability to convert 1’s and 0’s into radio waves and then back into 1’s and 0’s. But there are three big differences:
1. WiFi radios transmit at higher frequencies. IEEE 802.11b and IEEE 802.11g radios transmit at 2.4 GHz, while IEEE 802.11a radios transmit at 5 GHz. The higher frequency allows higher data rates.
2. WiFi radios use much more efficient coding techniques that also contribute to the much higher data rates. IEEE 802.11a and IEEE 802.11g use orthogonal frequency-division multiplexing (OFDM) and IEEE 802.11b uses complementary code keying (CCK).
3. The radios used for WiFi have the ability to change frequencies. IEEE 802.11b cards can transmit directly on any of three bands, or they can split the available radio bandwidth into dozens of channels and frequency-hop rapidly between them. Frequency hopping allows dozens of WiFi cards to talk simultaneously without interfering with each other.
Because WiFi radios are transmitting at much higher frequencies and because of the encoding techniques, they can handle a lot more data per second than our walkie-talkie example. IEEE 802.11b is rated at 11 Mbits/sec, while 7 Mbits/sec is more typical. But if there is a lot of interference, IEEE 802.11b data throughput may fall as low as 1 or 2 Mbits/sec. IEEE 802.11a and IEEE 802.11g are rated at 54 Mbits/sec, while 30 Mbits/sec is more typical.DONNA BALLAST is BICSI’s standards representative, and a BICSI registered communications distribution designer (RCDD). Send your questions to Donna via e-mail: [email protected]