[LINK] TV Spectrum for Last Mile

[George Bray]

> So. Does anyone on Link care enough about the broadband-over-TV-channels
> idea to explain how it stacks up into a business case that genuinely
> competes with DSL etc?
>
> RC

I know we've sparred on this before Richard, but here goes. This isn't
a business case, just my understanding of the current opportunity to
use TV spectra for last mile delivery.

1) The early work on BushLan by ANU said they were achieving
100-200kbps. Many years ago now.

<http://wwwrsphysse.anu.edu.au/bushlan/background.html>

2) When I investigated this issue for the wireless broadband enquiry
my discussion with management of Broadcast Australia indicated that
yes, using the VHF band was certainly possible and there existed at
the time USB modems that worked to 400kbps. That technology might be
better now too. The upstream link still uses a copper modem line, at
whatever that will do.

3) We have world's best expertise in RF electronics in the CSIRO, so
it's plausible that IP over VHF/UHF could be developed further.
Perhaps specifically to make good use of the national BA network for
regional users.

Using the TV spectrum for a downlink has been diss'd on this list
before, based on the experiences of people trying to get DVB-T
television channels. Certainly, it's not ideal reception everywhere.
But where it does work, it works well. That digital TV channel you're
getting is up to 10Mbps coming down that antenna.

At this point in the debate, however, I'd say that IP over VHF/UHF is
a potential technology for regional networking (100km from town) that
needs to be considered as a solution for *some* last miles. Combined
with the nationwide resources of the Broadcast Australia transmission
network, linked to a national fibre network, using IP on the TV
spectrum can be a last mile where other technologies don't reach.

This technology only needs to compete with DSL if the organisation
funding it is Telstra.  Any other organisation assessing the business
value (in terms of the establishment cost, and end-user utility) of TV
spectra would be thinking about the broadband black holes it fills.

I firmly believe that the VHF/UHF network can contribute to more
diversity in the BB last mile solutions. If the regional copper
network is not capable of high downstream speeds, why not use existing
high-bandwidth, wide area RF?

I'll try and find Gerard at ANU and see where this scene is at.

George





-- 
George Bray - University of Canberra, Australia