Australian researchers from National ICT Australia (NICTA) have developed a lower power, short-range chip for wireless communications that can achieve up to 5Gbps -- allowing them to transfer a complete DVD in a matter of seconds.
In March last year, NICTA researcher Tim Walsh told ZDNet.com.au the NICTA chip development team was aiming for speeds five times faster than today's fastest wireless standards -- however, the researchers have achieved speeds closer to 10 times faster.
NICTA project leader Professor Stan Skafidas suggested another possibility for the chip, dubbed GiFi: "[DVD] kiosks that you walk up to and when you are near it you could download a movie in a few seconds and watch it on your video iPod or put the device near your plasma TV and beam the signal," he said in a statement.
Skafidas said commercial samples of the devices would be available in "roughly" 12 months, and that he expects to see the chip on the market within three years.
Skafidas said the chip was a result of three years research and AU$35 million in "in kind" funding, that in part came from third parties such as IBM. Twenty seven people worked on the project including 10 PhD students.
While the chip has a maximum transmit distance of 10 metres, Skafidas said that the optimum distance would be around two metres.
While the chip is short ranged, short range transmissions are more secure. When combined with their high speed, this makes the NICTA chip suitable for secure wireless data transmission, Skafidas said.
The NICTA wireless transceiver achieves its high speeds by taking advantage of the unlicensed, higher frequency end of the wireless radio spectrum: from 57 to 64GHz. This millimetre-sized part of the spectrum is less crowded than the 2.4 to 5GHz part of the spectrum used by regular wireless devices.
Skafidas said in an interview with ZDNet.com.au that the chip used a significant portion -- as much as 3GHz -- of the 57 to 64GHz high frequency spectrum to achieve these speeds.
The five by five millimetre chip is based on the CMOS (complementary metal-oxide-semiconductor) process, which is the current process for creating massive volumes of cheap silicon transistor devices.
Skafidas said basing the chip on the CMOS process was a significant milestone in reducing cost. Normal high frequency transmitter chips are based on the much more expensive gallium arsenide semiconductor material -- as it generates less signal noise at higher frequencies.
By moving from gallium arsenide to CMOS, Skafidas said costs for producing high frequency receiver chips could be reduced from "five dollars per millimetre squared to ten cents per millimetre squared".
According to Skafidas, creating their chip using the CMOS process required "a whole bunch of tricks", included putting many components on a single chip, and a "mixer with very high isolation". The mixer allowed them to overcome some of the signal noise associated with chips made using the CMOS process.
Skafidas declined to name any commercial partners, but said; "We have been approached by some quite well known names." Skafidas also spoke about sharp international competition in this area. "It's a race," he said.
The chip was developed by the Victoria Research Laboratory in Melbourne and NICTA's Australian Technology Park Lab in Sydney. International research giants including IBM, Princeton University, and Georgia Institute of Technology also collaborated in the project.