Comment: Plenty of room for NBN wireless and Telstra's LTE

You have to feel for Mike Quigley when dealing with the vexing questions posed by the growth of wireless broadband connectivity: NBN Co has to simultaneously rebut the suggestions that wireless is a threat to the case for fibre connectivity, but at the same time respond to concerns that fixed wireless for the first 4 percent outside of fibre range is not an "inferior technology."

The last week hasn't helped.

As expected, NBN opponents have seized on comments made in an otherwise glowing Greenhill Calliburn review of the NBN business case that highlight its dependence on assumptions around the proportion of wireless-only homes and average revenue per user. (The good news is that these are the two risks the business plan actually models in its sensitivity analysis.)

Then came the co-incidence of Telstra announcing it intends to build an LTE network and tests carried out by a number of other carriers.

Malcolm Turnbull speculated on radio that the NBN "certainly is less likely to be financially viable" following Telstra's announcement.

I say bullocks, for a number of reasons. The development of LTE (and future 4G) wireless is very much a complement to the NBN in metropolitan areas, not a competitor. In fact, high-speed wireless services will require the rollout of more fibre at the backbone.

Telstra certainly doesn't think their announcement jeopardises the NBN. Would they have announced it just as they are concluding a deal to receive $13 billion from NBN Co and the Government if they did?

Let's consider some other reasons:

1. It's in the plan
NBN Co's plan has always allowed for the deployment of 4G. Despite comments to the contrary, the plan does envisage a further increase in the number of wireless-only homes. The business case was made with this in mind.

2. Telstra's wireless build is in metro and mobile, NBN Co's is fixed wireless for the bush
NBN Co CEO Mike Quigley has stated that the NBN Co wireless solution would be an LTE solution without all the functionality that provides mobility, like inter-cell handover. NBN Co has been instructed by the Government in its Statement of Expectations to acquire spectrum but not to bid for the 700 MHz.

In that sense NBN Co wants to provide a fibre-like solution to those in the bush, where rolling out fibre is cost-prohibitive, but does not intend to compete with Telstra for mobility in the cities.

3. LTE needs more bandwidth for ubiquitous coverage
Telstra announced a metropolitan-only LTE build in the 1800 MHz spectrum, with the only target device being a USB modem from Sierra. At the same time NetComm announced an LTE WiFi router developed with Sierra.

Telstra's network vendor, Ericsson, led the lobbying for the allocation of the 2 GHz spectrum for 3G services in 2000, and for 720 MHz and 2.5 GHz for 4G currently. However their premier customer is deploying 3G at 850 MHz and now LTE at 1800 MHz.

While this should puncture some of the arguments the industry mounts through AMTA about the need for extra spectrum just to launch 4G, it doesn't change the problem the industry has with overall bandwidth.

While the peak speeds of 4G technologies are impressive, they aren't making the same gains in spectral efficiency. In fact wireless network capacity (bits per hertz) only doubles every two and a half years (Cooper's Law) while bandwidth demand tends to grow with processing speeds that double every eighteen months (Moore's Law).

That's why AMTA and its members have been lobbying for the release of the additional spectrum, not because it is technically required for 4G, but that it is required for the capacity that 4G demands. It is also why President Obama's much vaunted discussion of wireless broadband in the USA depends on the FCC identifying an additional 500 MHz of spectrum.

There is a third way to increase capacity other than improved spectral efficiency and more spectrum, which is simply more base stations and smaller cell sizes. But then we'd be better off with fibre to the home and WiFi. This is ultimately Mike Quigley's simple case against wireless.

On a related note...
The Greenhill Caliburn report did support my line of argument four weeks ago – that "NBN Co's Corporate Plan is what it would expect to see from an Australian blue chip company" with "the level of detail and analytical framework that would be expected from a large listed public entity."

Further, the report recommended the Commonwealth seek "detailed, periodic disclosures to the Shareholder Ministers to allow them to assess the implications of any variations [to the plan] on a more timely basis and evaluate and introduce mitigants as required" – precisely the same argument I made last year when discussing why NBN Co should conduct a Cost-Benefit Analysis.

Nice to have some validation!