On Thu, 12 Jan 2006, Mike Bristow wrote:

In article . 170,
Adrian wrote:

Oh, and you're underground, so wireless will be problematic.

Most people wandering around on the underground seem to manage to have
comms.

Analogue voice comms has a low-bandwith requirment, and is tolerant
of noise. Data often isn't.

Voice comms has a low bandwidth requirement? Since when? I think even
compressed, radio-quality audio needs something like 8 kb/s - not a
massive quantity, but still plenty for the kind of data ticketing needs to
move. The noise is also a red herring - yes, data is extremely sensitive
to noise, which is why electronic engineers have decades of experience of
moving bits over noisy channels without errors. If there's enough
bandwidth for analogue voice traffic, then there's enough bandwidth for
reliable data traffic at a similar rate - it's just a question of getting
the protocols right.

The wavelan at home has issues with the base station in the lounge and
the laptop in the kitchen... and that's a tiny distance compared to even
simple stations.

And working at much higher data rates, and on a different, and highly
interference-prone, frequency. Ever used a modern (ie DECT) cordless
phone? Those are digital, working at 32 kb/s, and they work rather well in
my experience. The DECT protocol stack even has provision for data
transmissions.

Of course, it is a solvable problem, but it's not trivial. Read
expensive.

Actually, i'd call it a solved problem, and therefore trivial and pretty
cheap!

All this aside: I don't work in the rail industry, so my guessess may
be very wrong. However, I do know enough that even aparently simple
things like this often end up being more complex and expensive than you
expect.

That, of course, is true.

tom

--
On two occasions I have been asked [by members of Parliament], 'Pray,
Mr. Babbage, if you put into the machine wrong figures, will the right
answers come out?' I am not able rightly to apprehend the kind of
confusion of ideas that could provoke such a question. -- Charles Babbage