Whither Wi-Fi?
As head of an influential standards group, Dennis Eaton is eagerly sought out by would-be Wi-Fi entrepreneurs for his stamp of approval. And with the wireless networking field in the middle of a tug-of-war, winning his backing has become more important than ever.
But though this Wi-Fi seal of interoperability is considered de rigueur for wireless networking, the industry is in the middle of a tug-of-war, says Eaton, chairman of the Wireless Ethernet Compatibility Alliance.
Some voices within the sector think 802.11b networks need to be faster. Others advocate slowing them down and thus ease their notorious ability to quickly drain a laptop, phone or PDA battery.
There's also confusion looming on Wi-Fi's horizon. By next year, products using three brand new wireless networking standards are expected to reach stores. That will leave it to the WECA to help decode the differences between standards with awkward names like 802.11i, 802.11g or 802.11a
On top of all this, Wi-Fi is facing a space crunch. Nearly 30 million of these networks are expected to be in place by 2005, all using the same crowded swath of spectrum. The hunt is now on for new areas that Wi-Fi can operate in.
Even WECA is in transition, expecting to change its name to the Wi-Fi Alliance, pending the appropriate approvals. The organization plans to implement a new way of certifying that wireless LANS based on the various 802.11 standards can work with each other.
Eaton discussed these and other issues during a recent interview with CNET News.com.
Q: Is anyone talking about making Wi-Fi faster?
A: There are efforts underway within IEEE (a standards group) to provide even higher data rates in the 2.4-gigahertz band. It most recently kicked off a study group to take data rates beyond 54 megabits per second. What this group will do is look at data rates above 100 megabits per second. The "Wireless Next Gen" working group at IEEE started to see industry consensus around a particular idea.
| Over the long term, we believe dual band will be the dominant technology out there. |
Why won't WECA certify it?
It's not an industry standard. For the most part, those are single vendor solutions.
Why is WECA just now beginning to certify 802.11a products?
We've been waiting for multiple vendor solutions to show up. We only had a single vendor out there for a year and a half or more, and since we certify interoperability, there's been no products to test it against. But most recently, we had other 802.11a products show up.
People are concerned that Wi-Fi is beginning to crowd out the unlicensed frequencies it operates in. Is it possible to find more space for Wi-Fi in other areas of spectrum?
I believe the next unlicensed band is at 57 gigahertz. As it turns out, its not a real good band for wireless LANS. It turns out that at 60 gigahertz, oxygen resonates, so signal level is much lower. There's been some talk of potentially freeing up other bands. But the FCC hasn't started any formal consideration. There's nothing specific there, but it's been a topic raised from time to time. For the foreseeable future, LANS are locked in at 2.4 and 5 gigahertz.
Some companies are expected to have .11g products on the market before the spec is signed off on. WECA apparently can't do any testing until next year, so some products will come out without Wi-Fi certification. How does WECA plan to handle the confusion?
I don't think it'll be any different than what we've seen on 802.11a, which was on the market prior to WECA starting certification. For consumers buying those products today, they are most likely getting them from a single vendor, so it may not represent a big interoperability issue. Where you might run into interop problems is when you get multiple vendors. But that's when WECA starts the certification. I don't see this as a "confusion issue" for the consumer.
There's A product, B products, G products. On top of that, there will be "dual products," like A/B and A/G. But WECA has decided to call all of these "Wi-Fi". How are you going to differentiate all these products with one brand name?
There's a lot of details I can't share with you because there's no formal announcement yet. We try to take a long-term view when studying branding solutions. Sticking with "Wi-Fi" for these new products made more sense. Then, we could use a checklist to show the capabilities of the product.
Do A products have a life on their own, or are they going to forever be coupled with B?
We've already got standalone A products, and we will continue to see that for some time. But over time, it just makes sense that the market would move to dual band as the economics of a dual band solution start to make it lower cost. Over the long term, we believe dual band will be the dominant technology out there.
WECA is apparently testing .11a products now--whose products are you testing? Atheros?
I can't share that.
What is the time frame for the tests on .11a?
| Where you might run into interop problems is when you get multiple vendors. But that's when WECA starts the certification. I don't see this as a "confusion issue" for the consumer. |
What about combination 802.11a and 802.11b products?
First thing you'll see certified are access points, because implementations are already on the market place. But you can just stick two separate radios in any access point nowadays and call it dual band. If you believe the press releases, you'll probably start to see those in early next year.
What is happening with .11e, the quality of service standard for handling audio and video transmission? Is that close to being published?
That's probably going to be published mid next year. You may see some things come out before the standard comes out. I don't think the standard's firm enough right now to release product, but it could be as we approach the end of the first quarter of next year.
What strides are being made now to help ease the power crunch WiFi takes on mobile devices? Any standards work being done there?
We haven't seen anything significant in the architecture of chips to reduce power. Embedded Wi-Fi is using firmware features that do the best they can by putting it in sleep mode, or shutting down parts of the radio when not being used. We should start seeing chipset vendors releasing brand new architectures designed to conserve power. But that's in the future.
What radical architecture changes?
For example, instead of creating radios constantly pushing data at the maximum range, they might design a radio that doesn't perform at the peak rates. That'll save power.
Won't it kind of kill Wi-Fi's edge over other home networking in the speed category?
As any technology advances, there are usually the trades that you have to make. Ultimately, you'll see trades in direction of lower performance. To put it into context, think about the PDA for a second. When you're going to be doing some surfing or slurping some e-mails off the LAN, you're not streaming a lot of data, you're checking your e-mail or a flight schedule. So you don't need that much bandwidth.
The federal government recently reviewed the security of LANs, giving what could be considered a rather negative review. But they do say that "the standards community is aggressively working towards a more robust, open and secure solutions in the near future." What's your reaction?
I would agree for the standards that are built into equipment now, WEP (Wireless Equivalent Privacy) aren't as good as it could be. A business or the U.S. government should not rely on WEP alone to serve as the sole security mechanism. In the near future, maybe the first quarter of 2003, there will be a new security standards available for Wi-Fi equipment built in to make it much more secure.
What is the status of that standard?
It's called 802.11i. They are very close to a draft standard. We're hopeful they will be able to start moving towards a ratified standard towards later this year.
