Kerry doesn't expect the amendment to the standard to be completed until 2005 or 2006. However, when the standard is done and dusted, Kerry says it will offer more than just a performance hike over the data rates offered by existing 802.11b (11-Mbit/s over 2.4GHz) and 802.11a (54-Mbit/s over 5GHz) standards.
This is because the study group is actually trying to increase the throughput of 802.11 -- the actual amount of data that is transferred over a wireless link in a set period of time --- rather than just upping the data transfer rates. Anybody that uses wireless LAN regularly knows that, although the data transfer rates of the existing standards look fine on paper, once a wireless link is established that connection can be subject to interference, lost packets, and all sorts of nastiness, so actual transfer rates tend to be much lower than advertised.
The High Throughput Group is trying to deal with some of these issues. "We're talking true throughput here," says Kerry. "We've had proposals running at 108 Mbit/s and on up to 320 Mbit/s."
The high throughput group is certainly not the only major tweak to the specification that the IEEE is planning. Kerry also gave a rundown of all the other amendments and best-practice documents that the IEEE is working on in relation to the 802.11 standard. There's plenty of alphabet soup to go 'round!
Table 1: Current 802.11 Working Groups at the IEEE
|802.11 Specification letter suffix||What it does|
|e||Adds quality of service features, multimedia support|
|f||Defining how access points interact|
|g||Ups data transfer rates on 2.4GHz band to 54 Mbit/s|
|h||Adding dynamic frequency selection to 802.11a to comply with European regulations|
|j||A version of 802.11a for Japan that will run on the authorized 4.9GHz-5GHz frequencies|
|k||Defines radio and network information to allow the better management of wireless LANs, will also enable new applications like location-based services|
|l||Not being used, because it looks confusing|
|m||Maintenance, this working group will go back and correct any errors in previous ammendents to the specification|
|n||Not official yet but likely to be the designation of a high-throughput variant of the standard.|
|o||Not being used, because it looks confusing|
— Dan Jones, Senior Editor, Unstrung