Broadband future: NBN should be wired, not wireless, says Quigley
- 10 December, 2009 15:20
- Comments 15
NBN Co chief, Mike Quigley
Fibre to the home will be the only way to deliver high end applications such as complete remote computing and full 3D virtual collaboration in Australia, according to NBN Co cheif, Mike Quigley.
Speaking at the ealising Our Broadband Future forum in Sydney, Quigley quashed suggestions that an NBN should include or consist of large segments of wireless broadband.
“While wireless is currently enjoying high growth rates it has inherent limitations compared to fibre,” he said. “It is simply very difficult to overcome the limits imposed by physics. Spectrum is a scare resource and there is just so much you can do to increase spectral efficiency using better modulation techniques and coding schemes.”
Check out the event photos in the Computerworld slideshow
Despite this, real increases in speeds and download capacities of wireless had been seen in the last decade, largely due to the increasing number of cell sites and the shrinking of cell sizes, Quigley said.
“So why can’t we continue to increase the number of cell sites and decrease the size of cells? We still have to haul traffic out of the cell sites and for that we need fibre,” he said. “As traffic demand increases we can decrease the size of fibre connected cells so we end up with a very large number of cells, each serving a small number of premises. But is that likely to be a cheaper option than fibre to the premise? I don’t think so.”
It was also important to remember that wireless was a shared medium, so contention had to be factored into the access traffic calculations, Quigley said.
“Peak speeds may be high and equal to fibre currently, but average speeds are dramatically lower,” he said. “It’s also the case that users at the edge of a cell experience a far lower grade of service than those at the centre.”
Quigley said a user at the centre of a cell in next generation wireless may experience speeds of 150 megabits per second, but at the cell edge the peak may be closer to 10 or 20 megabits per second. Speeds further decreased when multiple users were access connectivity within the cell.
“Despite of these limitations there is a place for wireless in the NBN world. One would be foolish to think otherwise,” he said. “One would be equally foolish to the think that wireless will solve all our traffic needs, particularly as high definition video and other high capacity remote applications proliferate.”
Citing Cisco figures, Quigley added that while global wireline IP traffic was expected to grow to about 55 exabytes a month by 2013, mobile IP data was only forecast to grow to about two exabytes by 2013.
In contrast, current global wireline IP traffic is about 15 exabytes per month against less than one exabyte for wireless.
Email Computerworld or follow @computerworldau on Twitter.
- Bookmark this page
- Share this article
- Got more on this story? Email Computerworld
- Follow Computerworld on twitter
- Reconciling Datacenter consolidation and security: It starts with an integrated approach
- Unified & Collaborative Communications
- Pathways Advanced ICT Leadership Development Program Brochure and Course Outline 2012
- Why Hackers have Turned to Malicious JavaScript Attacks
- IBM zEnterprise System Brings Hybrid Computing Capabilities to Midsize Organisations
-
20 popular Ubuntu Linux apps you may want to try
-
Nokia N9: Why you shouldn't buy this device
-
Microsoft at a loss over Event Viewer scam
-
Customer service still dogs Telstra
-
Customer service still dogs Telstra
-
Introduction to Statistics for Forensic Scientists
-
Writing Scientific Research Articles - Strategy and Steps
-
Proteins - Structure and Function
-
Statistics for Terrified Biologists
-
Molecular & Cell Biology for Dummies
-
Molecular Forensics
-
The Scientific American a Day in the Life of Your Brain
-
Fluorescence Applications in Biotechnology and Life Sciences
-
An Introduction to Molecular Biotechnology - Molecular Fundamentals, Methods and Applications in Modern Biotechnology
Comments
Anonymous
Mike Quigley is right. Wireless IP will have a very significant role in the future of Net access, but there is no way it will have the spectrum, bandwidth and latency needed for all or most IP connections in 10-20 years time, within the horizon of the NBN plans we need to be locking in now.
Frank
Wouldn't Fibre also be wireless (so to speak, it's glass or acrylic, no wire).
Anonymous
Actually, there is another problem with wireless: almost all providers concentrate on increasing download speeds. I'm connected with a number of providers (as well as Telstra) and typically it takes minutes to send emails with relatively small attachments (of order of a few MBs). For remote access upload speeds are much more important. Cable/Fibre is the only way to go.
Anonymous
@frank
But cables are made of
@frank
But cables are made of wire, and you can have a fibre optic cable.. IT'S A CONUNDRUM!!!
/facepalm
Raj Sharma
Bill Gates said internet was only a tool for hobby users and not commercially viable; therefore Microsoft will no invest in the internet!!..1996..
..I think there is a world market for about five computers" ..and so the prediction go!!
Lets see if Mike Quigley's forecast on Wireless Technology and future bandwidth proves to be correct. I am confident he is wrong!!
Ning
As a employee in RF industry, I would like to raise the health and environment issues of wireless technology.
Chris
why are people trying to debate that fibre is not wired... thats just someone waiting for an aurgument.. its interesting that an example of bill gates predicted use of the internet as come up.. however another issue that people are forgetting to through into the mix is the latency of wireless.. at our current moment in time.. the latency is to high for reliable real time usage..hopefully this will be overcome in the future
Adam
Fibre optic cables - being glass - does not classify it as "wireless".
The naive mentality that 'Wired' translates to metal cables/wiring is absurd. This shouldn't even be a part of discussion on here.
Wireless, like your mobile internet, has it's uses and applications, but will not provide as mentioned already in the article the high-speed and reliability of optic cabling.
Wired, or FTTP (Fibre to the Premises), is simply referring to the direct connectivity (like your phone line) to a building. That is all.
Anonymous
will fibre optics to the home
will fibre optics to the home need to be trenched with other underground services,and as Telstra's copper access is not available will this mean digging through driveways/gardens to get it into homes?
Anonymous
As a regional wireless ISP, I would like to humbly offer my comments. In Regional Australia, fibre is way too far off. A viable intermediate solution is wireless. Satellite is not an option, except as last resort. No amount of spin will change this.
But lets define this word "Wireless". there are two types, Mobile wireless which requires enormous resources, and What I term Carrier grade wireless - lean, fast and reliable.
Mike Quigley has it wrong on a number of issues. Firstly, there is a lot of spectrum around, more than enough. With current technologies, Carrier grade wireless requires a mere 30MHz and with TDD, Mimo, syncronisation, and wimax with NO extra-grid roaming ability, we can provide a carrier grade service complete with redundant paths.
Mike Quigley seems to be talking about mobile wireless, which is a best effort network. That is not carrier grade wireless, and does require enormous amounts of spectrum for the higher throughputs, with still no guarantee of performance possible. Congestion and cell edge issues will always constrain throughput through these cells. Mobile wireless can not be compared to a carrier grade wireless connection.
But best of luck anyway to the NBN bureaucrats trying to deploy mobile wireless as a fixed line replacement in regional australia. I can only hope that they will eventually drop a few crumbs off the table so that we can deploy these carrier grade systems in Regional Australia right now, and thus let the NBN get on with doing whatever they do best.
georgina_swan@idg.com.au
@Chris. Your point about latency is spot on. It was really interesting listening to BHP Billiton Iron Ore CIO, Brad Wearn, speak yesterday at the forum about the challenges of providing bandwidth in the Pilbara. Latency, he said, was the real killer, esp for satellite technology.
Anonymous
Raj,
Physics says he is
Raj,
Physics says he is right. Bill Gates wasn't talking about a physical law, he was talking about new markets. Bill Gates was trying to predict <i>human behavior</i> he wasn't trying to predict something that mathematics has already predicted.
Mike Quigley predicted that the world will only ever need <b>more bandwidth and less latency</b>. He said that wireless won't be as good at providing bandwidth or low latency. Now he wasn't saying: "Humans will never make wireless higher bandwidth and lower latency than what we have now".
He wasn't saying: "Humans will never make useful wireless access" (because that is blatantly false, wireless is useful right now. It just isn't anywhere near as good as wired solutions, and doesn't cut the mustard for many things you can do right now with wired solutions.
He *was* saying that wireless will never be as good as wires. _This has been proven by the Mathematics and Physics that all of your electronic devices adhere to._
Bill Gates was good at writing software, but bad at predicting humans (in the past). Quigley is predicting nothing, he is explaining that Mathematics says Wires are always better (which they are).
Merari Schroeder
Quigley displays here his complete ignorance of wireless technology. He only has omni-directional LTE in mind when he talks. There's also directional radio wireless and Free Space Optics. There is 300Ghz of radio spectrum and it's all being used very inefficiently today.
My post here goes into some numbers and looks deeper into the future:
http://forums.whirlpool.net.au/forum-replies.cfm?t=1280334&p=50#r999
Also, why is Quigley not building LTE as well as FTTP? Are all customers going to pay $100 for FTTP and then $50 for LTE? He'll find a lot will only pay for LTE - losing customers. If he also builds an LTE wholesale network he'll have double the customers - probably double the cashflow for only 5bn or (10-25%) extra.
Mike Parnell
There is a quote from the latest copy of Motorola's sales magazine from "Paul Steinberg, chief architect with Motorola"
"LTE subscribers will have this high speed data experience ... with a downlink speed close to six to eight megabits per second"
The sales people for LTE say 380 Mpps, the engineers say 8 Mbps.
It doesnt do anyones case any good by pretending a best effort network (LTE) is anything but a mobile limited network.
It will not save Regional Australias problem, just squander any bandwidth that is around.
Anonymous
Problems with wireless. Shared medium, latency and health issues with increased power (Microwave and beyond).
Fibre is the way to go.
Post new comment