Draytek Vigor 2110 Series User Manual page 48

while the MGCP protocol uses client-server architecture, the calling scenario being very
similar to the current PSTN/ISDN network.
After a call is setup, the voice streams transmit via RTP (Real-Time Transport Protocol).
Different codecs (methods to compress and encode the voice) can be embedded into RTP
packets. Vigor V models provide various codecs, including G.711 A/µ-law, G.723, G.726 and
G.729 A & B. Each codec uses a different bandwidth and hence provides different levels of
voice quality. The more bandwidth a codec uses the better the voice quality, however the
codec used must be appropriate for your Internet bandwidth.
Usually there will be two types of calling scenario, as illustrated below:
Calling via SIP Servers
First, the Vigor V models of yours will have to register to a SIP Registrar by sending
registration messages to validate. Then, both parties' SIP proxies will forward the
sequence of messages to caller to establish the session.
If you both register to the same SIP Registrar, then it will be illustrated as below:
The major benefit of this mode is that you don't have to memorize your friend's IP
address, which might change very frequently if it's dynamic. Instead of that, you will
only have to using dial plan or directly dial your friend's account name if you are with
the same SIP Registrar.
Peer-to-Peer
Before calling, you have to know your friend's IP Address. The Vigor VoIP Routers will
build connection between each other.
Our Vigor V models firstly apply efficient codecs designed to make the best use of
available bandwidth, but Vigor V models also equip with automatic QoS assurance.
40 Vigor2110 Series User's Guide