WTF is... WiGig
Wireless wonder to soup up networks, kill USB, HDMI cables?
Feature It's now more than two years since the Wireless Gigabit Alliance (WGA) released the first full version of its 7Gb/s would-be next-gen Wi-Fi technology. There's been some activity in the intervening 26 months, including the first big multi-vendor interoperability test, but the second of these "plugfests" has only now taken place.
So why is it taking so darn long to get WiGig, the standard the WGA is promoting as the future not only of wireless networking but of cable-free computing too, into our hands? We need it now: we're streaming more HD video, doing more back-ups over the network, swapping more files between laptops, tablets and phones than ever before.
WiGig emerged in 2008, arising out of work done during the first half of the last decade to devise wireless technologies for streaming HD video content from players to screens, many of them based on ultrawideband technology operating in the 60GHz band: the section of the electromagnetic spectrum running from 57.24GHz to 65.88GHz.
Wireless HD? Why restrict it to tellies?
Why, said the minds behind WiGig, should we limit such technology to streaming video between set-top boxes and TVs? What about all the other types of data consumer electronics kit, computers and phones share? And why limit it to speed sufficient for compressed 1080p video and multi-channel audio?
The upshot of such conceptual thinking: Wireless Gigabit, a networking technology that extends Wi-Fi into the 60GHz band to ramp up speeds to a theoretical peak of 7Gb/s, for a multi-antenna, multi-carrier set-up, or 4.3Gb/s for single-aerial, low-power applications. It would incorporate Wi-Fi for compatibility with existing wireless networking software stacks, combined with a 60GHz PHY for the actual transmission.
In May 2009, the WGA was formed to steer the specification's development and promote the finished product. By the end of the year, it was promising a Q1 2010 release of the WiGig 1.0 . It was late, as these things often are, but not so very much: the WGA published the spec in May 2010.
At the same time, international standards body the IEEE agreed to use WiGig spec as the basis for its 802.11ad 60GHz networking standard, what will undoubtedly be the next generation of Wi-Fi.
The technology isn't perfect, of course. While 60GHz operation makes for much faster speeds than the 2.4GHz or even 5GHz bands can provide – higher frequencies for speedier per-frequency transmission, and a much wider frequency range to ensure more data can be transmitted in parallel – the range falls significantly too. Down to a few metres in fact, though the WGA says that WiGig's incorporation of beamforming techniques allow it to reach beyond 10m.
WiGig devices will be able to negotiate a specific beam path and tune their directional antennae accordingly. If someone interrupts the beam – no, 60GHz signals won't pass easily through people – the devices quickly pick a different path, relying on a nearby wall to reflect the beam around the obstacle.
Room to roam
Getting a signal through impermeable walls is another matter. Handy, then, that Wi-Fi is part of the scheme, since it provides a fall-back network when a connected device moves out of line-of-sight of the base-station.
Unconcerned by distance considerations, VESA, the organisation behind the DisplayPort digital monitor connectivity standard selected WiGig as the basis for a wireless version of the computer-to-screen system , which the WGA duly announced, in November 2010, as WiGig AV.
Since then, the WGA has published WiGig 1.1 and extended the specification to allow it to operate as a transport for PCI Express and USB traffic, as the WiGig Bus Extension (WBE) and the WiGig Serial Extension (WSE), respectively. WiGig AV became WiGig Display Extension (WDE), adding HDMI support  in the processes.
Look, ma: no USB, HDMI or DisplayPort cables
It sounds like the ideal wireless data technology. Support for high-speed networking combined with built-in compatibility with the existing de facto standard – which also handily allows it cope with its limitations – and it's able to operate as a wireless carrier for a variety of cable protocols.
It's WiGig's Protocol Adaption Layers (PALs) that allow it to operate as a wireless replacement for a variety of wires, from USB to DisplayPort and HDMI. Since it'll carry PCIe traffic too, it could even form the basis for wireless Thunderbolt. Range is not an issue here and neither is bandwidth. WiGig supports device-to-device connections, so there's no need to router cable-replacement traffic through a base-station.
So why, two years on from the releases of WiGig 1.0 don't we have it yet? Ali Sadri, an Intel boffin who happens to be the WGA's Chairman and President, admits the development of the hardware that will deliver the standard is "absolutely taking longer than it was originally thought".
Getting an entirely new radio system to work is not easy - and that's before engineers then have to make sure it adheres to the formal specification and works with devices from other vendors. Much of the plugfest activity in the past year or so has centred on that kind of testing.
"This is all-new technology," says Sadri, "not an incremental update to an existing standard. Every piece of the specification needs testing. We've had to proceed slowly, step by step. WiGig chipsets are not simple; we couldn't have shipped them this year or last year."
The technology sports a new MAC, he says, designed to be "more TDMA-like" (Time Division Multiple Access) than Wi-Fi has been, a trick learned from the cellular and WiMax worlds, and with quality-of-service and low-power operation at the heart of it.
Ultra wireless for Ultrabooks?
But progress is being made, he insists. WiGig 1.2 and the final version of IEEE 802.11ad should both be published by the end of the year, paving the way for the first commercial devices to debut in 2013. Other WiGig proponents, such as chip maker Wilocity, say the same thing.
That said, it's going to be 2014 before the WiGig begins to achieve critical mass as more devices ship with tri-band Wi-Fi radios: 2.4GHz, 5GHz and 60GHz. UltraBooks in particular will favour the technology, says Sadri, though as an Intel guy you might well expect him to say that. iHS IMS Research, a market watcher, reckons that by 2016 up to 100m WiGig chipsets could be shipping, 20 per cent of them tri-band products.
It will be Wi-Fi that drives the shift to 60GHz. The Wi-Fi Alliance, the body behind the popular wireless networking standard, will have decided how to bring 60GHz operation into the fold by then. It's currently promoting 802.11ac, and 802.11ad, whether it's ratified this year as Sadri forecasts or next, is an surefire technology for the WFA to adopt.
Evolution of Wi-Fi
Bringing 802.11ad into the Wi-Fi specification doesn't mean embracing WiGig entirely, only its data networking component. Sadri and many other WiGig backers hope the WFA will incorporate more of Wireless Gigabit, to allow Wi-Fi to evolve into a broader wireless technology.
It's already moving in that direction: technologies like Apple's AirPlay and Intel's WiDi (Wireless Display) are showing users that Wi-Fi is a technology that can do do much more than connect computers. While 802.11n made these technologies possible, 802.11ac gives them extra bandwidth to really stretch their legs.
But it can't ever become a true cable killer. WiGig can, insists Sadri. Wi-Fi, as it stands, ties clients to an access point; WiGig enables direct device-to-device connectivity too, and that's not a part of 802.11ad. WiGig delivers 2-4Gb/s with a four-element antenna measuring just 5 x 5mm, easy to put into a phone, a TV, a set-top box – pretty much any device you can think of.
You might get 600Mb/s out of 802.11ac using a similar size antenna, but it would still cost you a lot more in power consumption, says Sadri. Comparable speeds will require such a bulky antenna array, and that it'll put mobile usage out of the question.
Still, there's no escaping the fact that 802.11ac has a part to play, delivering the beyond-the-room networking WiGig simply can't do. But then there's the stuff 802.11ac can't do either: it can't support an office full of laptops all streaming lossless video data – which you need if you're displaying sharp text not video, for which human eyes' blur tolerance is much, much greater – to deskbound displays and connect to dozens of wireless USB devices while they're at it.
As Wilocity CEO Tal Tamir puts it, WiGig is placed to "totally eliminate the need for cords and connectors, enabling virtually any mobile device – such as a tablet, notebook, Ultrabook or smartphone – to act as a hub for high-speed connectivity and data transfer".
That's the potential and, with the last-minute polishing going on, this time it actually looks likely to be delivered. If we get the chance to rid ourselves of sync cables, it'll have been worth the wait. ®