[LINK] How fast is the NBN?

[Paul Brooks]

On 1/03/2016 8:57 AM, Tom Worthington wrote:
>
> On 28/02/16 13:46, Frank O'Connor wrote:
>
>> So all the observed trends, the increase in speeds ... aren’t gonna
>> appear...
>
> Speeds will increase, but people want stuff they can carry around with
> them, not have it stuck on a desk at home.
Only the devices I choose to carry around. I want my TVs, security alarm system, audio
system, kitchen appliances stuck at home - and operating connected 24x7, so a docking
station system where everything stops when I take the CPU away with me in my pocket
need not apply. They are also likely to require far more capacity and speeds than any
stuff that is small and light enough that I might carry around.


>
>
>>> Cell phones were invented to overcome the limited spectrum.
>>
>> ... I’m interested in knowing EXACTLY what you were trying to say.
>> ...
>
> Cell phones use radio transmission in small geographic areas, called
> "cells", which allows the spectrum to be reused. The cells were
> originally many kilometers, but now can be tens of meters (for
> example covering a few houses in a street):
> https://en.wikipedia.org/wiki/Cellular_network#Frequency_reuse
When cell-sizes are many kilometres radius, covering areas of up to 100 square
kilometres, it is efficient to build the high cost of the central radio transmitters,
receivers, masts and mountings, power connection and fibre or microwave backhaul
connection, as this cost was very small in comparison to running a real link to each
of the locations within the area, or compared to the large number of mobile devices
and people that could be served in that area and the costs amortised over.
When you shrink the cell-size, so that much the same capacity and radio bandwidth is
shared amongst a much smaller set of users, and you then need vastly greater numbers
and costs of basestation infrastructure that efficiency evaporates at some point and
it becomes more costly (and more prone to failure) to build a gazillion base-stations
than to just dig and run a cheap cable.

Also, in the context of frequency re-use, the signal strength at the edge of the zone
has to be weak so as not to interfere with an adjacent zone using the same
frequencies. This means there is a band around the circumference where the signal
strength is measurable, but too low to provide satisfactory service - yes the signal
extends many kilometres, but the outer half-kilometre is useless. In a large cell this
zone is a small proportion of the total area, so is acceptable. As you shrink the cell
diameter, that 'useless zone' becomes a larger and larger proportion of the total area
of the cell, and the cell becomes less space-efficient.

A cell with only 'tens of metres' range, covering a few houses in a street, might
provide sufficient signal strength at the street to provide great service to cars and
all the letterboxes at the bottom of the driveways - but that's not where people need
service. At the actual dwellings at the end of 5- - 15 metre driveways, and with
signal levels dropping further through the walls, is precisely where the signal will
be weakest and slowest - the useless zone.

Paul.