NICTA's GiFi chip to reach the market next year

By Liz Tay on Feb 24, 2011 6:18 AM
Filed under Hardware

Establishes company in the US.

NICTA invited the public to its Techfest at the Australian Technology Park on Wednesday. ...	More than 30 research teams were represented at the event, demonstrating biomedical implants, ...	NICTA's 60GHz gigabit wireless (GiFi) chip promised to deliver tri-mode wireless connectivity at ...
See all pictures here »

Confidential agreements between NICTA and major consumer hardware vendors could bring its gigabit wireless technology to market by next year.

In development since 2004, the 60GHz ‘GiFi’ chip promised to deliver tri-mode wireless connectivity at 10 times the speed of today’s 5GHz 802.11n technology.

A WirelessHD-compliant prototype was demonstrated in February 2009, and boasted audio and video transfer rates of 5Gbps.

The research team has since improved the technology’s robustness and compliance with the WiGig standard, researcher Jerry Liu told iTnews today.

NICTA hoped the technology would remove the need for cables for transferring large amounts of information between devices such as PCs, laptops and high-definition monitors.

GiFi was a hundred times faster than Bluetooth, NICTA claimed, and a single CMOS chip came at a tenth the cost of current technologies.

NICTA spin-off Nitero – established to commercialise the GiFi chip – joined Samsung, Intel, Cisco and Dell as a member of the ‘WiGig’ (Wireless Gigabit) Alliance.

Nitero was headquartered in the US to target the larger market. Three staff members were located overseas, and 10 others worked on improving the technology from Australia.

Liu said GiFi chips would be on the market by 2012 through agreements with smartphone and tablet vendors.

The project was expected to have cost a total of $10 million, and involved IBM, Synopsys, Cadence, Anritsu, Agilent Technologies, Ansoft, Conventor and SUSS MicroTec.

7 comments in this discussion

"@Martyvis "For wireless to be feasible (getting return on investment from your tower/cell/base station) you want to share this say 25Gbps amongst maybe 50 to 500 users." The technology in the ..."

By MerariSchroeder

Tags

nicta, gifi, intel, wireless, wifi, bluetooth, tablet, smartphone, nitero

NICTA GiFi spin-off gets $1.43m in Federal funding

RIM now giving away PlayBooks to admins

Ballmer eyes 500m Windows 8 sales in a year

Hungry Jacks public wi-fi goes national

Breaking Stories

CommBank suppliers compete for portable workloads

New virus hits Iran computers

Exetel taps artificial intelligence

Hackett steps down from Internode

ACCC approves Optus NBN deal

Readers of this article also read...

Comments: 7

MerariSchroeder Feb 24, 2011 1:32 PM	Wow, there seems to be a flood of wireless news. Too bad it's all fairly old (in terms of technology capacity). I had a look a while back, and it appears that even the antenna is MEMs packaged in side the chip (one of those figure 8 directional antenna). That's how they achieve the much lower cost. The signal is directional and is expected to bounce around the room many times before reaching a receiver. I could be wrong but I also remember seeing, that the antenna employs beam forming. Allowing the direction of the signal to be changed by circuitry (no physical movement required). Of course it goes without saying, that this reinforces my claim that wireless is not a dead-end. Bluetooth and WiFi are omni-directional radio systems. It employs directional technology to gain more speed (rather than focusing on spectral efficiency).
MerariSchroeder Feb 24, 2011 1:35 PM	The technology is typically limited to a room or house because of attenuation in the air. But imagine what could be achieved with the same techniques for macro-scale connectivity. It's conceivable that within the next 10 years (before the NBN is finished) we'll see the ability to have last-mile 10Gbps wireless connectivity. What will the future bring for wireless? We certainly cannot be pessimistic.
martyvis Feb 24, 2011 3:49 PM	I better get in before all the anti-NBNers do. Just looking at http://en.wikipedia.org/wiki/WirelessHD you can see that in this 60GHz band, you might get to 25Gbps (only 5Gbps today as above). But remember that this is one connection making use of the whole 7GHz wide band. For wireless to be feasible (getting return on investment from your tower/cell/base station) you want to share this say 25Gbps amongst maybe 50 to 500 users. All of sudden it brings you back to only to be able expect say 50Mbps per user (during peak time). This is fast, but still nothing like getting say 100Mbps to 1Gbps per user out-of-the-box with the GPON technology being rolled out by NBN. And when WirelessHD is ready and able to deliver 25Gbps, that same fibre infrastructure will fairly easily upgrade to something based on the 10Gbps fibre tech that is deployed in already in almost every data centre today. Yes we want good and fast wireless, but it still does not supplant the need for hardwired broadband services.
anonymous Feb 24, 2011 4:56 PM	@Merari "What will the future bring for wireless? We certainly cannot be pessimistic." The future will certainly see many improvements in wireless tech, as always. It will continue to be a useful adjunct to a fibre network. And the future will see even more improvements in FTTP tech, because as most people here recognise, there is far more upscaling possible without the spectrum, latency and reliability problems that restrict wireless networks.
Mordd Feb 24, 2011 8:02 PM	God getting in before all the NBN anti-fanboi's as well who will claim this is the death knell for optic fibre cable. Martyvis sums it up pretty well though so I will save you the rant other than to say I would love to see this technology trialled in a dense metropolitan area with tens of thousands of users all connecting at the same time in the same small area to see exactly what the effects really are and how much it actually slows down. My guess would be users will find they are no better off than with 10/100 network cable speeds once everyone was connecting at the same frequency in the same area x 10,000+ users.
umbria Feb 25, 2011 8:50 AM	Wireless is great and getting better, but its physical capacity will remain a constraint causing congestion and dropouts until fibre is widespread. The solution is not more towers, but sharing the load from mobile users by roaming them to WiFi where available. In a few years all but 7% of buildings will have fibre and most will share it with WiFi. Mobile devices that are in our near WiFi-enabled buildings (i.e. almost all of the time) then get cheaper data, lower latency, and higher reliability and speed than on wireless broadband. Why carry expensive water in a bottle when you are near a tap?
MerariSchroeder Feb 25, 2011 9:05 AM	@Martyvis "For wireless to be feasible (getting return on investment from your tower/cell/base station) you want to share this say 25Gbps amongst maybe 50 to 500 users." The technology in the article is directional. @umbria "Wireless is great and getting better, but its physical capacity will remain a constraint" The technology in the article is directional. Is seems that people don't understand the difference between directional and omni-directional wireless. When talking about omni-directional mobile wireless internet, you talk about spectral efficiency for the cell. When you introduce directional, the whole game changes. You can reuse the full spectrum among a plurality of paths. Nicta have packaged directional into an economic form factor for the living room. It's only a matter of time before directional is employed for the macro scale.