CSIRO’s hopes of commercialising its Ngara fixed wireless broadband technology for use in the National Broadband Network (NBN) may be hampered by radio spectrum recently purchased by the network wholesaler, researchers have warned.
NBN Co last month spent $120 million to purchase the entire spectrum portfolio owned by pay TV operator, Austar. The buyout provided access to the 2.3GHz and 3.4GHz bands for use in deploying fixed wireless broadband to four per cent of Australian premises not covered by the fibre-to-the-home (FTTH) rollout or remote satellite coverage.
The wholesaler is yet to publicly choose a fixed wireless broadband standard, with no indications as yet whether it will use existing LTE or WiMAX technologies, or another standard.
CSIRO ICT director, Dr Ian Oppermann, has previously confirmed negotiations with NBN Co over potential use of the research body’s prototype Ngara wireless technology, developed on the premise of utilising analogue television spectrum left vacant by the digital switchover in coming years. The technology proposes to provide committed symmetric speeds of 12 megabits per second (Mbps) to 1000 premises per cell area.
Representatives from NBN Co and the Department of Broadband, Communications and the Digital Economy, are known to have attended previous demonstrations of the Ngara technology and were in attendance at further demonstrations held at CSIRO this week.
However, the project’s manager, David Robertson, said the spectrum currently owned by NBN Co wasn’t optimal for deployment of Ngara technology.
“We could use the 2.3GHz system - there’s nothing stopping from reconfiguring the system to do that,” he said.
“But our view and our coverage maps would indicate that for the four per cent which the government has defined needs to be covered by wireless, our view is that the 2.3GHz coverage area is a lot smaller than you’d get with a 700MHz spectrum system and with our beamforming technology.”
CSIRO first trialled its broadband technology outside the NBN-connected town of Smithton in Tasmania on 14 December last year, after initial delays from a September timeframe. The trial reached expected speeds of 12Mbps upstream and downstream to six simultaneous users at ranges of between 10 metres and up to 16 kilometres.
The next stage, soon to be trialled in NBN mainland site Armidale, will use a frequency division multiplexing technique - similar to that used by several LTE developments in Australia - to attain speeds of up 50Mbps symmetric for four users, or 12Mbps to up to 48 users.
Further refinements are expected to reach a total of 1000 users using several base stations per area.
“What we’re trying to show with this system is that it is scalable... just apply contention ratios of 20 to 1 and you get your 1000 residents that can be covered,” he said.
On both occasions, CSIRO has used seven to 28MHz portions of the 600-700MHz spectrum under a scientific license from communications regulator, the Australian Communications and Media Authority (ACMA).
Robertson said prototypes of the technology currently utilised wideband components to suit several different bands within the licensed spectrum, as final allocations would not be known until the product is commercialised. However, he said the use of existing television broadcast towers was a vital aspect of the technology being developed by CSIRO.
“It isn’t just putting the pole into the ground - you’ve got to put electrical equipment there, you’ve got to get power to that tower, you have to overcome any objections from residents,” Robertson said. “We’re using broadcast transmission towers which are already there.”
Negotiations have also continued with other vendors to potentially commercialise the technology. In either case, however, the research body has targeted the technology at covering the “digital donut” of premises falling outside NBN Co’s fibre deployment; an issue Robertson said was already evident.
“We saw that at Smithon - the mayor who lived 200 metres up the road couldn’t get it because they didn’t roll fibre out that far,” he said.
“Most of the high disposable income people tend to live in the digital donut area - why live in a high density part of town when you can live on acreage outside of the town.
"That explains why, in my view, why the take-up has been low in some of these areas. They put fibre in areas that aren’t matching the demographic that actually meets the fibre.”
Follow James Hutchinson on Twitter: @j_hutch
Follow Computerworld Australia on Twitter: @ComputerworldAU