60GHz: the gigabit wireless we are waiting for

Send to a friend Print

Help more people find out about this story

Del.icio.us
StumbleUpon

Seamus Byrne20 August 2008, 10:49 AM

Straight out of Intel’s research lab comes the first public discussion of 60GHz, or mmWave, wireless networking. Want 3.8Gbps over air? Just wait until the next Olympics.


802.11n might only just be joining the mainstream, but Intel is already hard at work on bringing is wireless that will offer modern wired equivalent speeds over short range networking. 60GHz, or millimetre wave (mmWave), networking is the latest target for breaking current wireless bandwidth barriers.

The 60GHz range has already been in use as an outdoor and space-orientated networking technology, but due to oxygen attenuation it has been seen as too power intensive to be used as an indoor solution. Until now.

The brains in Intel R&D (and other work by teams at our own NICTA and CSIRO) have been finding there is great short range potential for 60GHz when the antenna are built into MIMO-like arrays and given directionality. The promise right now is data rates of 0.9Gbps over 16m, 1.9Gbps over 7m, and 3.8Gbps over 4m, but tests have shown upwards of 5Gbps at even shorter ranges. This puts 60GHz on equal footing with USB3.0 (4.8Gbps) and PCIe2.0 (5Gbps) as a potential ‘last metre’ solution for high-speed wireless networking.

There are also a number of interesting side effects of the directional antenna requirements. Some are negative, with line of sight a requirement and a problem with any obstructions, even people, potentially breaking communications. But the aim is to work on redundancy through reflected pathing to create what amounts to an ideal single room network solution.

The highly positive aspect is improved performance when multiple devices are in use. Whereas current wireless technologies share bandwidth, decreasing overall capacity per device as the frequency traffic grows, directional use of 60GHz means more devices can ‘talk’ at full capacity. Whether in the home or the office, Intel R&D Manager Lily Yang calculates there are 2x and 3x performance gains available in high traffic situations.

Yang hopes to see 60GHz reach the mainstream in four to five years. This may arrive sooner in a number of forms, with short range ‘Sync & Go’ applications a likely first stop. There is short term promise too in home theatre applications such as wireless display technology, with wireless computing and LAN applications the later arrivals.

Between now and the arrival of 60GHz as an available product we have plenty of new wireless toys to play with, from wireless USB to UWB, but the bandwidth potential of 60GHz makes this one wireless technology to wish for as part of your next-next computer network.

Seamus Byrne travelled to Intel Developer Forum San Francisco as a guest of Intel.


Post your comment



Comments

RSS feed Email alert

GoughLui (Cornerstone member):

I think this is potentially quite expensive to implement given the high frequency and high power required to overcome attenuation. It possibly won't be suitable for battery powered devices. Even wireless N is having trouble with adoption due to cost ... I can't see the change to 60Ghz to be any better. Sure there is higher bandwidth, but there's possibly other spectra which are more manageable than 60Ghz. Somehow I think Intel just did this as research, its practicality is questionable. After all, if it can affect air molecules, what would it do to our cells?

20 August 2008, 1:23 PM (10 months ago)report abuse Send to a friend reply

gankul (Cornerstone member):

It will be interesting to see the final fruits of intels research into this, it might make it only a big corperate enviroment technology, who knows, maybe router reflectors on the ceiling? :p

20 August 2008, 1:55 PM (10 months ago)report abuse Send to a friend reply

LostBenji (User):

The main reason for the push to 60Ghz is the recent world realease of that spectrum under the "Free License" banner. mmWave is an interesting area to experiment in but it has no penetration through objects at all. It can only travel throuigh aout 0.2mm of skin. Where it does come good is that almost anything will reflect some of the energy making it better to bounce for NLOS (Non Line Of Sight) links. The limited range due to attenuation of the air will suck.

Lets face it, if you need to be a meter away for the speed then why not just use a cable and keep it simple... ?
For RF links over range and efficient data transmission then go for lower frequencies and give up on Gb speeds.

21 August 2008, 7:26 AM (10 months ago)report abuse Send to a friend reply

anonymous user Anonymous user


Tags