[LINK] pCell speeds

[Jim Birch]

A white paper on the technology, named DIDO last iteration before
rebranding.

http://www.rearden.com/DIDO/DIDO_White_Paper_110727.pdf

On my reading, this looks like a good, even radical, technology because it
overcomes the problem of interference by treating all antennas on the band
as an array and so works with interference rather than treating it as a
hard limit to antenna density.  It squeezes a lot more data transmission
out of limited spectrum.

However, it's a bit oversold or, surprise, under-explained in the press
articles.   What they don't mention is that the system needs to add an
antenna to the array for each receiver, or rather, for each additional
100%-of-bandwidth user.  If these users are doing HD video chat you
probably need (like) an antenna per user, but for lower bandwidth
applications you would get a lot of punters per antenna.

These antennas can be low power - and thus much cheaper - because the user
receives a part on his composite signal from each of the antennas in the
array and because they can be closer.   However, all the antennas must
connect to a single big data centre that would be doing some serious real
time Fourier-type number crunching to compose/decompose the signal to/from
each antenna.  I wonder how the
multi-antenna-multi-receiver-moving-client-changing-transmission-environment
maths problem scales?  The connection between the antennas and the data
centre will require some mysterious as yet unknown technology - or
alternately a fibre network.

Since phone frequencies have wavelengths of the order of 10 cm it should be
possible to place antennas in clusters on a rooftop to reduce the costs (I
think).  It should also allow low-cost antenna drops into black spots and
inside buildings without screwing up the rest of the suburb.

The low power antenna and reduced spectrum would lower the cost of entry
into the cell network business, especially if there's an existing fibre
network to plug your antennas into.  There are some interesting business
models there.   The system would work with existing phones/protocols but
I'd guess that there would be advantages in new phone protocols down the
track.

I don't see this as removing the need for a fibre network but it could
postpone or even eliminate the looming max-out time for cell networks.  If
it flies, it certainly hits one of the big problems of cell networks.

You might get a version of this on your home wifi router before long too.
Come to think of it, probably sooner.

- Jim