Kevin P. Fleming
kpfleming at digium.com
Sat Jun 4 20:44:07 MST 2005
Tilghman Lesher wrote:
> Might be helpful for someone to summarize exactly how GR303 is
> similar or different to other signalling protocols (i.e. line
> appearance, callerid, did, etc.). Google doesn't seem to find any
> really specific information about it (other than it existing). For
> example, is GR-303 similar to a PRI? If so, how is it different from
> a PRI?
GR-303 is a concentration protocol, used to service subscribers over a
link similar to a PRI.
The equipment we use at the trade shows has a T-1 link on the GR-303
side, and multiple T-1 links on the "traditional" side. Those T-1s are
then connected to standard FXS channel banks, with analog phones
connected to them.
This allows the single T-1 coming out of the Asterisk box to serve more
than 23 end stations, although not more than 23 simultaneous calls,
obviously. As I understand it, GR-303 also supports "bonding" multiple
net-side T-1s together, so it could conceivably be used to handle a
large number of end stations off of 2-4 T-1 links.
GR-303 has value when you want to service 'occasional use' subscribers
(residential, or analog stations in lightly used office situations)
without having to have a dedicated T-1 channel for each subscriber
directly off of your Asterisk box. It may also have some value when you
have limited cabling resources between the Asterisk box and the location
where the subscribers are (like many miles of crappy copper <G>) and
can't run enough T-1s to handle them all directly.
That's the extent of my knowledge... I haven't actually seen GR-303 get
configured anywhere, I think Mark and Matt Fredericksen are the only
ones who have worked on it very much.
More information about the asterisk-dev