2 October 2007 12:22PM
Tags: cover | feature | wireless | broadband

CRN asks the question to industry experts, could the enormous anticipated cost
of a fibre to the node roll-out be avoided with wireless?

In the 10 years since the first public offer for Telstra shares launched on September 29, 1997, the Federal Government has tried to introduce a higher level of competition in the Australian telecommunications and data access markets, with some degree of success. We now have a handful of strong, competitive mobile phone carriers, good competitors in some business access sectors and backhaul providers, and an increasing number of ISPs providing access by installing their own DSL equipment in Telstra exchanges.

But the market for competitive data communications has and always will be hampered by our reliance on the aging copper infrastructure that provides the last mile access loop to our homes and most business premises.

A good reliable technology that continues to serve this country well, DSL over copper wires has one critical weakness that will never be overcome despite advances made in future standards. Physics dictates that no matter what you do, the signal will degrade over long distances, so those located close to the DSLAM will be able to connect at higher speeds than those further away.

Of course these problems are exacerbated outside the metropolitan areas where the large distances and low population density that make Australia the wonderful country it is, actually works against our new needs of easy access to high speed data access.

Federal intervention
The Federal Government has attacked the problem on two fronts. One set of schemes to assist rural and regional Australia is already in place and now metropolitan Australia is potentially in for similar treatment. Significant amounts of public money are likely to be invested in providing broadband access to those in the cities who can’t get it and faster access for those who can.

The trouble is there is no clear solution. Generally speaking, our low population density assures there is no business case to provide high speed access to regional Australia. Many regional ISPs depended on the now defunct Broadband Connect funding to survive. We have already seen in our Broadband feature last month, that the Government’s Connect Australia funding will be applied to a new access and backhaul network from wholesale service provider OPEL, a joint venture company formed by Optus Communications and regional services and ISP company Elders.

The only sector able to reliably secure funding outside the OPEL joint venture has turned out to be the satellite providers, which have secured funding to provide remote Australians with access under the new arrangements. Others, according to reports, are desperately pursuing their own exit strategies.

The trouble for Australian Internet users is that there is no real business case for additional infrastructure in the cities either. All the options are unpalatable to those charged with making a profit from shareholders funds, so it seems likely that the government will have to step in to provide a circuit breaker in terms of funding or regulatory changes to make truly high-speed broadband a reality for the majority.

BigAir chief executive, Jason Ashton, believes that before we can correctly find the answer we have to make sure we understand the question. That’s not as simple as it seems. To begin, there’s no clear agreement on what sort of bandwidth we need. Shara Evans, managing director of the Telco analyst firm Market Clarity is a strong supporter of fibre to the node (FTTN) or to the premise to enable users to quickly shift what she anticipates will be comparatively huge data files 100+MB in size. Others are more modest arguing that even if such large files are used outside of the business sector, it’s likely they can be sent piecemeal as required, heavily buffered, or as a background operation.

While it’s easy to see that video production houses, or radiologists might need to transmit hundred plus MB files on a regular basis, a service capable of providing this level of access may be less of an imperative in the home. But that’s the trouble with innovation, points out David Kennedy, Ovum’s research director for telecoms.

“The difficulty is that there is no real business case at the moment to deliver such high speed broadband into metropolitan areas, because there is still no clear picture of what service might be delivered over such a platform,” says Kennedy. “The very nature of innovation means that you don’t know beforehand what services might be required and which might be a success.”

Admittedly, there are some pretty easy guesses. Content services such as IP-TV and movies on demand both require significant amounts of broadband, but certainly not at the gigabit level for the foreseeable future.

However, if you step back from the question a little, there are a number of options to provide such services and part of the problem may be looking for one solution to realise everybody’s data dreams in one hit rather than looking for multiple solutions for multiple needs.

To attain Gbps speed to the premise you must install fibre all the way. The FTTN proposals so far put forward would still rely on VDSL or later DSL incarnations from the node to the home, and so it would still require Telstra’s copper local loop to the house. The huge expense of a fibre to the home (FTTH) deployment virtually eliminates it as an option in the near term, leaving FTTN as the only realistic option inside the next ten or twenty years. So firstly, would-be service providers need to build a business case to get FTTN and then further down the track will need an entirely new one to get FTTH.

Why don’t we all go wireless?
There are other technologies which could provide adequate data speeds for most users in the mid-term and some resellers are rightly asking the question, “Why don’t we all go wireless?”

Telstra has certainly been talking up the capabilities of its Next G network, all other mobile phone carriers have 3G networks and there are numerous pre-WiMAX and WiMAX-like wireless networks already being deployed.

We have seen a willingness by the government to fund wireless as a DSL alternative with the OPEL network, which will deploy WiMAX across unrestricted spectrum under a class licence and for the most part true WiMAX spectrum is underutilised, if used at all, in this country. Why not accelerate the roll-out of these networks rather than undertake the huge expense of building a FTTN network that we aren’t even sure we need? It is a fair question and one that may never be addressed thanks to the political climate surrounding broadband.

The reality is, however, these options present their own problems. Kennedy agrees that with a billion dollar funding grant, Unwired could use its spectrum to deploy a metropolitan WiMAX network capable of serving the needs of the community. But he cautions that the solution would present its own problems.

Despite WiMAX’s prodigious data capabilities, we would require a significant number of installations to achieve the cell density necessary to serve such a large population of users with high speed access. And that’s not to mention the backhaul problem.

Even Unwired’s chief technical officer Eric Hamilton hesitates at the idea. Although we didn’t offer him a billion dollar funding grant to provide ubiquitous WiMAX for every Australian home, he did mention that fibre is expected to cost $2000 per household while with $40 million he could cover the same area. Hamilton’s $40 million network would not provide the sort of base load data carrying capabilities we need though, and would act as only part of the solution with DSL playing a part.

To date, Unwired has been holding off on the deployment of mobile WiMAX, though it has committed to begin a roll-out next year.

Coverage, capacity and cost
Hamilton says you need to evaluate the three C’s when looking at wireless networks - Coverage, Capacity and Cost. “You want to get the maximum coverage and the highest capacity at the lowest cost,” he says. “WiMAX is closer to solving that equation most effectively”.

He’s certainly more hesitant about a FTTN network, pointing out that by 2011 there will be more people connected to the Internet wirelessly than are currently connected by all methods. Part of that growth will come from developing markets where wireless broadband is their first ubiquitous solution.

Unwired’s initial roadmap for WiMAX is to deliver just 3Mbps, but the standards process is already working toward 70Mbps and as Hamilton puts it: “The sky’s the limit.” WiMAX may not be the panacea we all hope for anyway. As Ovum’s Kennedy cautions, there are some doubts about WiMAX’s in-building reception and although the technology is promising, it is not yet proven: “We would all be a lot more comfortable if we had some large-scale real-world deployments of the technology.”

Other options we have seen investigated overseas more than in Australia include low frequency networks such as the 700MHz spectrum soon to be auctioned in the U.S. This sort of technology could prove to be a great solution in regional Australia as the signals travel very long distances, but without much data carrying capacity, making it suitable only for low-density populations. Like municipal Wi-Fi, another idea that has received much press attention in the U.S. is 700MHz spectrum networks that are perfect for addressing the type of digital divide issues faced by the U.S. Government, rather than capable of providing high-density populations with super fast broadband.

Although Australia has some municipal Wi-Fi plans such as the NSW State Government plan to blanket Sydney’s CBD with a free public 802.11 network, for the most part it’s a difficult solution to realise. The cost of access points is extremely low compared to WiMAX or 3G base stations, but the coverage is tiny in comparison and would require the vast sprawling Australian cities to deploy thousands upon thousands of access points. The solution doesn’t really satisfy Hamilton’s third “C”, capacity, as even with mesh topologies, the backhaul could quickly become a problem.

It’s not even clear whether there is a satisfactory business case for Municipal Wi-Fi. The deployments planned for the U.S. have hit a stumbling block with the main provider Earthlink cutting half its employees and scaling back its plans, cutting most city’s muni-Wi-Fi dreams off at the knees.

In one of the few cities to achieve the goal, Philadelphia, EarthLink initially estimated it could blanket the city with a network between 20 and 25 nodes per square mile. It then raised that number to 30 before finally being forced to install up to 47 nodes per square mile.

The cost proved prohibitive and potential service providers are now insisting that city councils sign on as anchor tenants to insure an income stream. The idea is not completely dead, but it’s bleeding.

The 3G card
That pretty much leaves the generation 3 GSM technology as a wireless broadband offering. Already widely available in the Australian market, 3G HSDPA is already reaching speeds of up to 7.2Mbps. What remains to be seen is how scalable this technology can be in the medium term.

Hamilton argues that 3G cannot match WiMAX for efficiency. As you would expect he is fairly dismissive of 3G describing it in terms we are sure he doesn’t want reprinted here. In short, he accuses it of being a data solution bolted onto the side of a voice technology, and even its long term roadmap will take years of development before it reaches the present day capabilities of WiMAX.

There’s little doubt that 3G can go a long way toward satisfying the needs of ad hoc and mobile users, but it’s unlikely to truly resolve the broadband dilemma faced by metropolitan users.

Mobile phone companies are pushing the solution
with data cards and USB modems, and like its competitors, NetComm has just released a 3G compatible wireless router, the N3G001W.

It’s the first device of its kind to be compatible with all the available 3G networks in Australia, and it even works with the iBurst wireless broadband network from Australia’s Personal Broadband.

“It’s particularly appealing to those people who are on the fringe areas of the ADSL network,” said David Stewart, managing director of NetComm. “These people will now be able to receive a reasonable if not higher-quality data network where they haven’t previously been able to get ADSL. It will also appeal to renters who have not previously wanted to pay the costs and make the commitment to ADSL. Of course it will also appeal to users in the huge bush market who are still waiting for any form of broadband.”

Ultimately, NetComm plans to release a device that will act as a 3G home gateway similar to the wireless broadband modem released in July by Virgin Mobile. This could become a popular solution for some. Optus, through its subsidiary Virgin Mobile, was the first to enter the wireless broadband market with a solution targeted at a DSL alternative.

Virgin Broadband provides a home gateway device which uses the 3G HSDPA network rather than fixed phone lines to provide connectivity eliminating the need for line rental.

It’s a neat idea for renters and people who can’t get DSL, but a far cry from providing the sort of true broadband that could dissuade users from FTTN with VDSL.

The question that remains is whether the luxury of fibre is something we can live without.

Written by Adam Gosling

This article appeared in the 1 October, 2007 issue of CRN magazine.