Eric, I do not think anyone can underestimate the need
for testing, verification and rewriting code, the
cooperation needed to go from switched ciruit
to packet networks, switches, controllers,etc,etc..
Especially the "cooperation" part when many handsets
by different manufacturers roam in even more numerous
networks, of diverse age, etc..
I'm also sure that the best solution for some, medium
aged "masts" might be to install only the minimum
to first have at least couple of channels capable of
doing all the new tricks the correct way.
(the oldest ones are easier to just replace 100%)
All depending on the modularity, upgradability
of existing, old boxes and protocols, capability
to fix, upgrade old code or reroute around it.
That is, nothing new for especially wired operators,
having gone from analog, relay switches to more and more
digital, 64kbps, ISDN and now different breeds of DSL
directly hooked up to interenet.
But the subscriber copper wires have a special room for
cross connecting wires and equipment, into the DSLAMs,
switches,etc and the handsets do not roam around freely,
do not jump over to the neighouring, competing operators
network every now and then (covered by earlier roaming
agreements for switched circuit GSM)
Must be fun and busy out in the roaming wireless world..
(luckily modern basestations, etc are easier to
"upgrade" with code than using the screwdriver on
older, but the worst is probably to start a campaign
of replacing roaming handsets)
Ilmarinen.
P.S. Spent one summer job in one of the most "diverse"
switches in Finland, equipment spanning at least
4 generations (long time ago but invaluable
experience) |
