Jump to content
The Inquirer-Home

Post-802.11n WiFi targets 1Gbps

Analysis Newly ratified standard promises more robust links and better mobile support
Friday, 18 September 2009, 13:28

NEXT GENERATION 802.11n WiFi products should be better than current 'Draft 11n' devices, even though these are compliant with the standard that was finally ratified on 11 September after nearly seven years of development and industry infighting.

Belkin launched a 'pre-11n' access point a full five years ago but the 600 or so 'Draft N' products on the market comply with Draft 2.0 of the specification that was approved in March 2007. This became a de facto standard three months later when the governing WiFi Alliance offered tests to certify that a 'Draft N' product would work with other conforming devices.

The industry is playing down differences from the official standard ratified by the IEEE organisation, and its essential mandatory features are indeed virtually identical with the draft standard. But the final standard includes several options either not implemented within current products or not covered by the earlier interoperability tests.

The WiFi Alliance will begin testing for some of the optional features early next year, enabling their use in next-generation 802.11n products.

The biggest gain, according to Richard Edgar, WiFi product manager in the handset business unit at Cambridge Silicon Radio (CSR), will be better support for devices like mobile phones that are too small to have more than one antenna.

He explains that when the WiFi Alliance approved the 802.11n draft, under pressure from users anxious to make early use of emerging Multiple In Multiple Out (MIMO) technology, it targeted the market that was likely to take off first. "What it wanted was the stuff you would need for a PC or laptop to talk to an access point."

This involved the use of various combinations of two or three transmitters, receivers and antennas. MIMO can send a different data stream on the same frequencies from each antenna, boosting throughput without increasing the bandwidth.

Mobile phones and other small devices don't have room for more than one antenna, which is why none of them support 'Draft 11n'. But options in the ratified standard, particularly a technique called Space-Time Block Coding (STBC), allow single-antenna devices to take advantage of MIMO.

This is particularly relevant to CSR, a leading supplier of combined Bluetooth and WiFi chips for handsets. It already has a chip that supports STBC but this is of little use until access points support the feature, and that won't happen until interoperability tests are available - probably next year.

Other features awaiting testing should boost the robustness and performance of all 802.11n links. They include:

  •  A ‘green field' mode that allows 11n devices to talk to each other faster in the absence of devices using earlier versions of WiFi, by cutting out signals telling them to shut up.
  • Packet aggregation. The use of larger data blocks reduces the overhead of wrappers such as headers and error-check codes.
  • Three data streams per MIMO link. Current tests cover only two and the standard allows for four, but no devices yet support that number.

Rachna Ahlawat, VP of marketing at Meru Networks, says the use of 'Draft 11n' is already widespread in corporations but she expects a lot of companies still using older versions to move on. "The useful life of wireless LAN equipment is about three to five years and 2005 to 2006 were big years for WiFi shipments. So many are coming up for upgrade anyway."

She points out that 802.11n is faster, with better management facilities, and more secure. "Basically if you don't deploy 11n you are going to have to worry about security."

In fact the IEEE has also just also ratified a second WiFi specification known as 802.11w, which plugs a potential security hole and will be incorporated into 802.11n devices. And that is by no means the final ingredient of WiFi's notoriously complex alphabet soup, which has run out of letters and is beginning to double them up.

Tasks groups are drawing up specifications for WiFi in vehicles (11p) and mesh networks (11s). One called 11v for wireless management could help Internet service providers minimise contention between their customers' WiFi networks, says CSR's Edgar. Another called 11z proposes a wireless 'tunnel' for pumping high-speed data between two devices, such as a phone and a TV.

But of widest interest are two known as 11ac and 11ad. The first is targeting throughput of at least 1Gbps using the 5GHz range of bands. This is more than three times faster than the rated maximum for an 11n channel - real throughput is of course lower when you factor in network overheads.

The 11ad group is looking at gaining the same kind of speeds using the 60GHz band, which is at the centre of a classic industry tangle. The band is already being used in proprietary implementations to push video streams from set-top boxes and recorders to TV sets, and two industry groups are already pushing competing standards.

The range at 60GHz is very short as the frequencies happen to be absorbed very easily by oxygen, but this is actually an advantage because it avoids the kind of contention between neighbouring networks that is already a problem with WiFi in cities.

The 11ac and 11ad specs are both scheduled for approval by 2013 but Edgar reckons this is optimistic. "I reckon the date will slip by a couple of years. Remember the 11n spec took seven years."

The 60GHz technology will be targeting many of the uses envisaged for ultrawideband (UWB), which lost some credibility when Intel dropped the technology last year and the governing Wimedia Alliance wound itself up.

To complicate matters even further, the alliance passed the UWB baton to the Bluetooth SIG, which was considering using the technology as a fast data pipe - a task for which it is also adopting WiFi.

Edgar does not think UWB is out of the picture yet. "There are already products on the market; 60GHz will take a few years. I think the two will end up being complementary," he said. µ

 

 

Share this:

Comments
Good article

Now that is a fine, technical article. Congrats.

posted by : Matias, 18 September 2009 Complain about this comment
OH GOD!.....

I'm gonna Need LOTS & LOTS MORE TinFoil!!!

Has, or is anyone involved with all this microwave stuff gonna bother to test what biological effects it is going to induce in Humans(yeah us squishy ones with the brains that overheat and get brain cancer if exposed to microwave signals) or other life forms?
maybe before they push even more of this gear on us and give us ALL! cancer...
I'm sure the NHS would like a heads-up so that they can prepare to increase funding to the MacMillan Trust etc.. for all those who will have been given Terminal Cancer....

posted by : Mord., 18 September 2009 Complain about this comment
Definitely worth reading

Good article - I commend you. I learned a few things today on the N standard and will be armed with these when searching for an upgrade to my non-MIMO Wireless G access point. This is the first good article I've read on the INQ since Charlie left. Good work. Keep it up and I may actually add the INQ back to my regular reading.

posted by : Max Weber, 18 September 2009 Complain about this comment
Fix 802.11n first!

This is getting ridiculous; 54Mbit devices provided about 24 Mbit/s throughput half-duplex. Ok, so over 50% speed reduction, fine. But what's with this "11n 600Mbit/s" bullshit that might get you UP TO 85 Mbit/s @1m/3ft distance if you're lucky and the stars are just right?

Lets see these 600 Mbit/s devices perform a bit closer to spec first and THEN talk about next generation wireless. If you could run 2.4GHz together with 5GHz and bond both links, sure you might get 150 Mbit/s, but has anyone seen anything like that available in real life (outside specialized business solutions that cost $5000+)?

Are we going to get 1 Gbit/s device that in reality manages just 100 Mbit/s?

Wireless industry stinks.

posted by : Jokes are also wireless, 18 September 2009 Complain about this comment
Waste of time

Why should I read an article which begins with crap like Gbps?

posted by : Gbit/s, 18 September 2009 Complain about this comment
Because its not crap

Gbps means the same thing as Gbit/s and is simply an alternative abbreviation following a different convention/standard. There are multiple standards and conventions for abbreviations so simply choosing one (such as Gbit/s) and stating it is more correct than other accepted conventions (such as Gbps) is meaningless.

posted by : Tavi, 21 September 2009 Complain about this comment
Oh rly?

Will that mean that i can finally stream those laughable 5MB/s needed for FullHD over 10m? Cause the 'n standard certainly can't seem to do it, even though i am supposedly connected with 300Mb/s.

posted by : Raven737, 22 September 2009 Complain about this comment
Advertisement
Subscribe to the INQ Newsletter
Sign-up for the INQBot weekly newsletter
Click here to sign up Existing user
Advertisement
INQ Poll

Christmas computer sales

Will you be buying a new computer this Christmas?