Blowing the digital dividend on wireless NBN

Blowing the digital dividend on wireless NBN

Summary: With the 2013 federal election likely to fall within weeks of the opening of the 700MHz spectrum, could Malcolm Turnbull again be considering an increased role for wireless?

SHARE:

With the next federal election still over a year away, Malcolm Turnbull has found great joy in leading the telecoms community in a prolonged game of Twenty Questions when it comes to the Coalition's alternative NBN policy. That has left speculation rife about what the plan will entail — for example, that the Coalition would actually buy Telstra's copper network outright — but a cursory look at the calendar suggests that something else may also be on offer.

Specifically: it cannot have escaped Malcolm Turnbull's notice that the next federal election will be held in late 2013 — and that this date coincides nicely with the end of analog television on 31 December 2013, which will enable the release of the precious 700MHz "digital dividend" spectrum. While Turnbull has kept details of his alternative NBN sketchy — apart from regularly repeating claims that it will use a fibre-to-the-node (FttN) architecture, and be cheaper and completed faster than Labor's plan — it's worth considering whether the digital dividend might also play a role.


(Cell tower image by Joe Ravi, CC BY-SA 3.0)

The Coalition's love for wireless has been well documented (and, indeed, past Communications Minister Helen Coonan recently revived the ghosts of the failed OPEL wireless project) but OPEL's reliance on the interim WiMAX technology was one of many problems with the plan. Another was its requirement for base stations on every few street corners to match demand with available 3G and WiMAX spectrum.

If, however, an alternative NBN architecture could tap in to long-term evolution (LTE) services running over newly freed 700MHz spectrum — which has been demonstrated to provide healthy speeds at long distances — theory at least says that a heavy wireless component might deliver decent speeds to large swathes of the population without the ridiculous density of base stations calculated under previous coalition wireless plans.

After all, commercial mobile operators are already planning to roll out LTE networks that will largely utilise their existing towers rather than requiring massive numbers of new sites. Using this for an NBN alternative, rather than distance-constrained WiMAX, might be more palatable than any wireless solution discussed so far — and would indeed get rural residents connected quickly and relatively effortlessly; that's why NBN Co is using it for the much-discussed 4 per cent of homes that are too remote for fibre, but too urban for satellite.

Theory at least says that a heavy wireless component might deliver decent speeds to large swathes of the population without the ridiculous density of base stations calculated under previous coalition wireless plans.

If such an infrastructure were available on a widespread basis, the Coalition could adjust its overall NBN costs by playing with the sliders on the network's coverage figures — balancing the capital costs of its own network against the cost of modifying or cancelling existing fibre NBN contracts. By adjusting Labor's 93 per cent fibre and 4 per cent LTE coverage to, say, serving 60 per cent of the population with FttN and 37 per cent of the population with LTE wireless, Turnbull could theoretically defer significant construction costs while quickly servicing broadband blackspots with something that's better than, well, nothing.

All of the usual concerns about wireless capacity and infrastructure would of course still apply, as would the need for extensive fibre backhaul to support the wireless infrastructure. But such a model would also call for a fundamental shift in the country's wireless infrastructure: since the 700MHz spectrum is expected to be auctioned off later this year, a new coalition government would need to negotiate with the eventual spectrum holders to gain a guaranteed swathe of its spectrum, presumably on a long-term lease arrangement.

Turnbull has already flagged this sort of arrangement when discussing satellite services; in his attack on NBN Co's $2 billion satellite program in February, Turnbull said the Labor strategy ignored the possibility of leasing what he said would be adequate existing satellite capacity (it's not, but that's another story).

Turnbull has made a habit of referring to privately owned networks as though they are indiscriminately available for use at the Coalition's whims. He has referred to Telstra's copper network many times in a way that suggests that Telstra was a willing and eager participant in years of fierce negotiation with Labor; he has also made a habit of speaking paternally about Telstra and Optus HFC networks, treating them as open and available to competitors — even though they're, well, not.

So, it's not a huge stretch to posit that Turnbull's faster, cheaper broadband solution could be based on configuring NBN Co as a mobile virtual network operator (MVNO) of sorts — taking advantage of its massive bandwidth demands to chisel bulk discounts out of commercial operators.

Those operators might be happy to have guaranteed business, but such a plan would also run the risk of saturating their bandwidth. LTE may be more efficient than WiMAX or 3G HSPA protocols, but once you throw in the demands of millions of mobile broadband users, you're looking at a significantly higher usage profile. Mobile telcos would need to do extremely careful usage monitoring and traffic shaping to ensure that their pursuit of government business didn't interfere with their own commercial LTE plans.

Were this to form a part of the Coalition's strategy, the government would need to offer carriers other sweeteners to ensure their participation. I'd suggest that this would likely take the form of extensive subsidies to help private operators build out their LTE capacity, particularly in rural areas where carriers can't just leverage off of extant mobile towers. This would return the country to the days of the government-run Australian Broadband Guarantee program — and it would be entirely consistent with the Coalition's subsidy-focused mentality.

It's not a huge stretch to posit that Turnbull's faster, cheaper broadband solution could be based on configuring NBN Co as an MVNO of sorts — taking advantage of its massive bandwidth demands to chisel bulk discounts out of the commercial operators.

Without more details from the Coalition, this discussion is purely academic and theoretical, but the coincidental timing certainly merits contemplation. Building a baseline capacity using wireless would indeed deliver on Turnbull's promise to broadband many areas faster and more cheaply than digging fibre — although the logistical issues and massive financial impact in leasing and auction-revenue opportunity costs would not be insignificant. Whether Turnbull would risk building a new strategy around wireless, when the last one failed so spectacularly at the polls, is another big question, but until he provides more clarity around the party's real plans, questions are all that we can really have.

What do you think? Would a 700MHz strategy deliver on Turnbull's objectives? Could the private-sector LTE industry service the capacity demands of such a solution? Would it? And should it?

Topics: Broadband, Government AU, NBN, Tech Industry

About

Australia’s first-world economy relies on first-rate IT and telecommunications innovation. David Braue, an award-winning IT journalist and former Macworld editor, covers its challenges, successes and lessons learned as it uses ICT to assert its leadership in the developing Asia-Pacific region – and strengthen its reputation on the world stage.

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.

Talkback

53 comments
Log in or register to join the discussion
  • Certainly an interesting concept David and the sort of thing the opposition should be suggesting rather than the old trusty, NBN = waste, white elephant etc...

    But in the end it still falls well short of the NBN, imo, because of the limitations (remember 93% of all downloads are still done via fixed - imagine a big block of that suddenly going via wireless) and particularly as $'s will be thrown at private companies for them to own and operate our network. The opposition complain about NBN waste, but the NBN has a return for the outlay, there's won't, if subsidies are paid.

    I think by now, most people understand that wireless/fixed are complementary. More people use wireless gadgets (my family has about 10 wireless gadgets and one fixed, for 4 people) but the bulk of the work is still performed by fixed... as per the ABS stats

    http://www.abs.gov.au/ausstats/abs@.nsf/Lookup/8153.0Chapter7Dec%202011

    We have always said the NBN is needed because private companies won't invest. But companies are already willing to invest in complementary LTE/Wireless, as they don't have an incumbent who already owns absolutely everything to compete with, as they have had with fixed. Ah remember the HFC roll out games and $b's wasted in competitive stifling?

    Then we still have the need for Telstra's (old, unkempt, obsolete) copper for FTTN, for which a deal with Telstra would need to be struck (as well as a deal for HFC with Telstra and Optus). Plus let's not forget FTTN "will need upgrading" and there isn't a clear upgrade path, which is why the panel of experts said FTTN WAS NOT viable about 5 years ago, and there is a very sizeable cost involved there.

    Let's not forget too, if FTTN/LTE isn't complete or not commercially profitable (private enterprise won't accept NBN like 7%) and our network is privately owned, what incentive will there be for them to want to upgrade to FTTP anyway? We will be back to where we are now, they will only invest where profitable?

    Look, whilst certainly an interesting idea, all of these grey areas have been done, or achieved, or eradicated etc, with the current NBN. An NBN we Aussies will own and will be a valuable asset for our future.
    Beta-9f71a
  • Sounds like the plan is to divert NBN cash into subsidising wireless technology that industry would build anyway. So in say 6 years time when the limitations of wireless are sorely apparent and its Labour's turn back in power then the fibre starts to roll out again. So total bill = current NBN cost + wireless subsidies + contract default costs + administrative overhead of this nightmare. And with that we end up where we would have anyway, just years later. That's some fine fiscal policy right there.
    xBeanie
  • I'm sure the NBN co. and their contractors have covered their back sides should the Coalition try to pull the pin on the NBN.

    It's going to be too expensive to undo (and pay out the penalties, like $500 million to Telstra, for breaking the contract they have NBN co. or slowing down the roll out: http://m.zdnet.com.au/telstra-nbn-co-finally-come-to-agreement-339317292.htm) the contracts that have been signed. Ergo the only option for the Coalition is to honor and fulfill those contracts i.e. keep rolling out the NBN.
    Jingles-8366c
  • FTTN sounds "great" in theory when talking about this patchwork plan of theirs. The big problem is (and I'm sure the coalition clowns have not even thought of this) is if you cancel the rest of the NBN roll-out how do you decide who gets FTTH and who gets FTTN?

    Say there is a town of about 5,000 who are getting FTTH in the 3 year roll out plan and under the coalition patchwork plan they would have got FTTN instead. Meanwhile in another town of 50,000 they are not covered by the 3 year plan but end up getting FTTN instead since the rest of the roll out is canceled. I’m sure Turnbull and his apologists thinks this sort of idiocy makes sense but keep in mind they are the ones who flip-flop between saying "Cities should get the NBN roll out first to pay for the rest of the roll out!!!" and "under served regional areas should get the NBN first because cities have HFC!!!"
    Hubert Cumberdale
    • The solution is very very simple. Let ****s like you who think they need 100/40mbps connections PAY to have fibre brought to their home from the node. SIMPLE!!!
      Frergers
      • And who pays to the node brainiac?

        Oh we do, via subsidies to private companies and we get 0 ROI and no asset ownership, speaking of wasteful mismanagement and white elephants, FFS.

        Wakey, wakey, start thinking for yourself instead of having the Liberal Party do it for you, faithful sheep.
        Beta-9f71a
        • ....and who pays for the NBN your omnipotence? Oh, we do via subsidies to private companies and we get 0 ROI and no asset ownership... speaking of wasteful mismanagement and white elephants FFS.

          Wakey wakey, start thinking for yourself instead of having the socialists do it for you, faithful lemming.
          Frergers
          • You really have NFI do you?

            You are unable to comprehend the NBN and the differences between the Coalitions half baked plan, sad.
            Beta-9f71a
  • 700MHz band will already be sold off although not started before the coalition takes power, they would have to negotiate with the winners as they can't just take say 5MHz from each person as that would separate the band. I don't agree with the plan but the only option would be to negotiate with the license winners on building the network for them.
    Look at it from the network point of view.

    Why would Telstra have a bar of this, they will need/want the spectrum for their own network not some quick fix to try and save a few dollars.

    On the timing front, The senate will still be in control of the Greens/Labour party for a long time after the election so what can they do apart from whining is call a double disillusion.
    SW_Victoria
    • Oh, I think double disillusion is guaranteed under a Tony Abbott government...
      Gwyntaglaw
      • Can't wait !!!
        Frergers
  • I'm sure wireless technology has its place in todays society however I'm yet to see portable devices which can work like desktop devices do. Its simply not practical to think just because mobile phone use is increasing that people will do the same work they do on their desktops with their phones.
    The next flaw with this plan is the fact that the electronic's - the backbone behind this idea will be outdated in a very short time & need replacing again and again.

    Face it, who uses a 7 year old computer?

    If this network was to last 7 years it would be old - outdated and due for replacement. Not only part of it, but the whole thing.

    The difference with fibre is - The fibre remains the same as its the backbone behind the network (possibly for more than 100 years) and when an upgrade is required you simply replace the media converters or lasers at the end of the network. The rest of the network is passive. The same fibre is used at 10 or 100 Megabit as would be used at 40 Gigabit (of cause with current technology a dwdm technology would be required for this speed)

    So what it comes down to is dollars and cents (or should I say sence) Lets look at the math.

    If you spend 17 billion on a network which will last around 7 years and then 7 years later have to replace that network including inflation @ around 4%Pa your 17 billion wireless network would end up costing another $21.5 Billion dollars PLUS the original 17 billion dollar investment is $38.5 Billion dollars by the end of 2020 and STILL not have the security nor be able to match the current speeds of the NBN

    Now thats a waste of money & this is not including buying Telstra's copper.

    This is bad management and a bad idea which will only cost the Australian people more in the long run with less return in investment, production and innovation than the NBN would currently bring.
    These FACTS only prove Tony Abbortt is looking after his own intest and does not care if the Australian population pays more - He just wants the PM's job.
    fibretech
    • Build a computer right, it should last 7 years. Mine is 5 years old, and its only now (yesterday in fact) that I looked at it and started thinking it needed upgrading - want more RAM, DDR2 is hard to find. But that was only because I planned for future needs when I built it. A couple of repairs along the way, again, as planned (extra HDD, updated graphics) and its still as servicable as the day I bought it.

      But it was through good planning, not luck. If I'd bought cheap at the time, I would have been upgrading years ago, which is the point your making. Take the cheap option, you're going to be replacing it sooner rather than later.

      What happens when 5G comes along? Do all the 4G towers need to be replaced? What about when 6G comes along to supercede that? Or 7G, 8G, and so on. Static technology is eventually going to be replaced, and for me thats what wireless represents.

      Copper and fibre at least have some room for growth.

      I guess for me its like comparing a desktop computer to a console. The console (wireless option) is a static option, and one you know will be replaced sooner or later. A desktop machine (fixed line) is fluid, and from the base model you can tweak it and modify it for years to come, giving that machine a much longer and more versatile life span.
      Gav70
      • "What happens when 5G comes along?"
        Not likely to have major speeds increases as Shannon is approaching rapidly. 5G will work more on smaller cells beater connect ability and general over all improvements but the industry does not know how to much more data than what’s proposed in LTE-A. Although there are ways around this it’s unlikely to effect the NBN too much as they have to go macro cells.
        SW_Victoria
        • Thanks SW_Victoria :) Point was more that the wireless network is a static piece of technology that needs to be replaced EVERY time a new technology comes along that delivers faster speed.

          You want that faster speed, you need to update every single tower - you cant run 4G on the 3G towers for example, which is limiting how fast 4G can be rolled out.

          The statement was to try and show that if we relied more on wireless, the rollout process would be repeated every time. Building expense time after time. I just used 5G etc because the sequential numbering was easy. The upgrading will happen regardless, but if there are more towers to upgrade, simply because there are more users, the overall cost spirals the longer you follow that path.

          NBN is essentially a 1 time rollout, with any future upgrades being done at the exchange, not the node or line. Much like upgrading DSLAMS now, that are upgraded to provide faster and faster services.

          The fundamental upgrade path for fibre is along the same lines, and as a result is a considerably cheaper option.

          To put it a different way, NBN is aimed to be rolled out by 2020. By 2025, there could easily be 3 or 4 upgrades in wireless technology that would see bulk replacements of connectivity needed.

          If 1/3rd of people relied on wireless as their sole connection, thats 3 times the current needs (at a guess - work with me), so 3 times the bandwidth needed for a consistent connection. And 3 times the upgrades needed every time...
          Gav70
    • Sorry I disagree with your statement about replacing the wireless infrastructure as the same applies to the Optical transmission, but I’m still for fibre.

      Let me explain: The Optical transmission requires a medium to travel through, in this case it’s a fibre optical cable. With wireless its nothing but air. So you never have to replace the air in between but you will have to replace the Optical (50-150 years, whatever) The equipment at either end will need to be replaced but the same also applies to any communication including optical.

      The PON initially will have a capacity of 2.5Gbps or thereabout per 32 houses or so, 2.5Gbps/32= 78Mbps. If demand rises above this mark or houses become more dense then the PON will need upgrading.

      Wireless, much harder to calculate.
      Let’s say 200Mbps per 20MHz with 4 channel is 800Mbps. (LTE)
      60 houses per cell.
      800/60= 13Mbps. (there’s the problem)
      SW_Victoria
      • SW - I agree with you about the meduim - air would not need to be replaced - nor would the fibre.

        The fact remains that the bulk of the cost of the NBN would be the installation of the PON or the fibre - whereas the bulk of Tony Abbotts wirless NBN would be the cost of the radio transmitter and recievers which would need to be replaced at much shorter inverval that does not represent good value for money.

        It is far cheaper to replace the media converters / lasers and ONT/NTU than it would be to replace all the electronic wireless reciever and transmitter equipment.

        I know I am preaching to the converted, but for those who don't understand from 23 years in the radio communications industry I can tell you that RF is not a good solution. RF or radio frequency has inherent problems that cannot be simply overcome, these include but are not limited to
        *Range - Distance from the transmitter
        *Lack of bandwidth and network congestion
        *Impluse noise - Lawn mowers, Planes flying overhead, Power tools
        *Cross Polarization - Rain Fade, Snow, Smoke from bushfires
        *Obstacles - Tree's, Mountains, Building
        *Atmospheric Conditions
        *Co Channel interference
        And the list goes on... Not to mention the 17 year old kid with his dual quad core processor sitting in his bedroom packet sniffing & cracking algorithms which in turn gives him access to your computer network (hacking)
        fibretech
        • anyone who has experienced problems with digital television reception can expect similar reception problems from a wireless NBN

          I was in Melbourne a couple weeks ago & took my wirless 3G 7.2 modem with me - The congestion was so bad I couldnt even get my mail till after 8pm as it kept timing out. Remember there is only so much available bandwidth.

          Fact remains Tony Abbott is only objecting to the NBN because he wants to make the current Government look bad so he can become PM - Even if this means the Australian public ends up paying more in the long run.
          fibretech
        • But... but... what about spiral waves? What about that nice Mr Perlman and his DIDO technology? What indeed about quantum entanglement communication? Those are surely just around the corner by now!

          AND WHAT ABOUT THE LASERS? Huh? Think we'd forgotten all about them, didn't you?

          :)
          Gwyntaglaw
        • Point is so does Optical, you can't just isolate wireless, but I also mentioned you will never reach the same speeds.

          "You want that faster speed, you need to update every single tower - you can’t run 4G on the 3G towers for example, which is limiting how fast 4G can be rolled out."
          Yes you can, Telstra does it Optus is planning to do it.

          "NBN is essentially a 1 time rollout"
          With constant upgrades for both Optical and Wireless stated in many press releases and interviews.

          "The fundamental upgrade path for fibre is along the same lines, and as a result is a considerably cheaper option." Can't agree for this one, definitely a no.

          "The fundamental upgrade path for fibre is along the same lines, and as a result is a considerably cheaper option." Only if aiming for the same benchmark which is my point.

          "To put it a different way, NBN is aimed to be rolled out by 2020. By 2025, there could easily be 3 or 4 upgrades in wireless technology that would see bulk replacements of connectivity needed."
          Not really as they all fall under the OFDMA technology bandwagon, the existing contract has the LTE-A in the contract which is not a major upgrade. What about upgrading PON's they aren't cheap.

          Final note: I'm for fibre but I don't want to pay for someone else’s fibre which the excess cost and blow outs if you go and build fibre to all the wireless location. Most of them are isolated so you would have to build trunk lines there. They are going down to towns without
          SW_Victoria