Government Computer News
Speed 'n' distance
The 802.11n standard isn’t yet final, but 4 devices show what it can do
THE 802.11N AMENDMENT to the Institute of Electrical and Electronics Engineers wireless networking standard was developed more than four years ago, but it has yet to receive final approval. Amid unresolved comments and delays caused by lawsuits over patent disputes, progress has been measured in the issuance of a new draft every few months. The current draft, 3.02, will not be the final version because there are still 127 technical comments outstanding. If we are lucky, the next draft, 4.0, soon to be put on the ballot, will be the final one, and we may see final approval by the end of this year. We at the lab, however, are not holding our collective breath.
Digg
De.licio.us
Slashdot
Technorati
Despite all this uncertainty, several brave wireless manufacturers have come out with versions of draft-n or pre-n access points (APs), following as closely as possible the draft specification as it is voted on and altered.
Some companies are waiting for the amendment to be finalized before manufacturing any 802.11n devices, and others are only developing routers.
If you are uncertain what improvements 802.11n might have compared with previous Wi-Fi versions, it can be boiled down to essentially one thing. The signal from an older wireless device — 802.11 g, b, or even a — must bounce around any metal obstacles until it reaches its target. But 802.11n devices have a smartantenna technology called multiple input multiple output. Three antennae talk to one another and switch off when one antenna’s bounced signal would interfere with another. This gives 802.11n devices an effective range nearly double that of a or b devices and a bandwidth almost six times that of g devices.
We collected draft-n access points from Belkin, Linksys, Netgear and SMC and prepared to run them through their paces. We decided to test access points as opposed to routers because we felt that the majority of our readers who were in the market for wireless would be adding that capability to an existing network with a router already in place rather than replacing that router with a wireless one. Once the 802.11n amendment receives final approval, this practice will certainly change.
We installed the recommended client adapter card on a laptop PC and made sure the two computers were on the same IP subnet as the AP. We created a folder of various types of files that totaled 100M of data, and we were ready to begin the transfer testing.
We were especially pleased with how easy it was to set up these devices. We didn’t have to do much beyond setting some IP numbers to get them ready to go.
We were also pleased to see that each of these devices has top-notch security. They all have Wired Equivalent Privacy, Wi-Fi Protected Access, WPA2 and service-set identifiers, along with 64-bit and 128-bit encryption. Fortunately, the standards that govern these things have been in their final forms for a few years, so everyone has adopted them.
All touted a data transfer speed of the 802.11n maximum 300 megabits/sec, but we knew we would be getting less. Although most of the APs had Gigabit Ethernet ports, we decided to connect them to a computer that had a 100 megabits/sec adapter, which is still the norm in most organizations. Also, TCP/IP slows transfer rates, because it sends data in packets that have to be checked for integrity. These two factors meant that we’d be getting results at a fraction of the theoretical maximum.
In laying out our course for determining transfer speeds at various distances, we decided to alter the conditions from last year’s test of 802.11n devices (GCN.com, Quickfind 1006), when we used a straight hallway with complete line of sight between the AP and client. This time, the client was in direct line of sight with the AP for the first 20 feet. From 30 feet to 40 feet, there was a single cubicle wall between them. And from 50 feet on, we went outside, putting a pane of glass between them in addition to the cubicle wall.
To get an average transfer speed that was as accurate as possible, we took timed transfers at each 10-foot increment out to 170 feet while moving the 100M file across the wireless network.
We threw out values that were significantly high or low, which would indicate some fluke. The remaining values were taken as an average. Each test site had at least three readings put into its average score.
Glass is supposed to be just as transparent to radio waves — of which wireless frequencies are a subset — as it is to visible light. However, we observed a drop in transfer speeds for some of the access points as soon as the laptop was outside the window or shortly thereafter. Of course, there are many factors that could have played a part in these drops, but we felt it was worth mentioning.
In addition to the transfer speed tests, we judged each access point on additional features it might have, such as multiple operational modes, additional Ethernet ports, the existence of a status display, and the variety of antenna deployment and mounting options. We also looked at the price of the devices compared with what you get.
Once the 802.11n amendment is approved, there will no doubt be an influx of APs and routers. But regardless of the results here, we commend these few manufacturers who are trying to stay on the cutting edge.
![1105 Government Information Group [valor]](/images/1105logo_website.gif "Government Computer News [valor]")
© 1996-2008 1105 Media, Inc. All Rights Reserved.
