RE: [EFM-P2P] RE: [EFM] T.V. broadcast / unicast
Let me try to clarify my definitions:
> b.1) Voice over TDM emulation, transparent mode.
> b.2) Voice over TDM emulation, advanced signaling.
> b.3) Voice over IP (soft switch approach).
(b1) is the simplest approach; a 64kbps channel is used to transport
exactly the same signal that is carried through the analog line. There is
no compression, nor special signaling (only simple bit patterns to
represent onhook/offhook and ring signals). As Bob pointed out, it should
work 'out-of-the-box' with any existing system, and it should not cause any
compatibility problems. I see no reason not to have this option available,
at least on first-generation systems; although it's clearly limited in
functionality, we expect it to be quite price effective for small scale
(b2) in this case, we are using advanced protocols such as V5.2 or GR303 to
interoperate with the PSTN. These protocols include signaling for special
services, and are used to connect DLCs to the central office switches.
Compression may or not be used. This is where our compatibility nightmare
starts, as every country has its own variant,and every vendor seems to
implement the standard in its own particular way (it remembers Tanembaum's
Law of Standards - there are so many to choose from...). Given that many
switch vendors are already migrating to IP-based systems, this option is
losing relevance (but it's still very important for today's market).
(b3) is the pure voice-over-IP approach. We expect to have fewer
compatibility problems in the near future with this technology; for now,
the situation is as bad as (b2) - or worse, given the relative maturity of
the traditional protocols when compared to the newer VoIP approach. The
main advantage for the EFM implementation is that it can be seen by the
transport layer as pure data; there is no need for special reservations for
TDM emulation at the lower levels.