Codecs; Jitter Buffer - Nortel BCM50 Installation And Configuration Manual

22 Chapter 2 IP telephone overview

Codecs

The algorithm used to compress and decompress voice is embedded in a software entity called a
codec (COde-DECode).
Two popular codecs are G.711 and G.729. The G.711 Codec samples voice at 64 kilobits per
second (kbps), while G.729 samples at a far lower rate of 8 kbps. For actual bandwidth
requirements, refer to the BCM50 Networking Configuration Guide (N0027156); note that the
actual kbps requirements are slightly higher than the label suggests.
Voice quality is better when using a G.711 Codec, but more network bandwidth is used to
exchange the voice frames between the telephones.
If you experience poor voice quality and you suspect it is due to heavy network traffic, you can
achieve better voice quality by configuring the IP telephone to use a G.729 Codec.
Note: You can change the codec on a configured IP telephone only if it is
connected to the BCM50, or if Keep DN Alive is enabled for an offline telephone.
The BCM50 supports these codecs:
• G.729
• G.729 with VAD (Voice Activity Detection)
• G.711-uLaw
• G.711-aLaw

Jitter Buffer

Voice frames are transmitted at a fixed rate, because the time interval between frames is constant.
If the frames arrive at the other end at the same rate, voice quality is perceived as good. In many
cases, however, some frames can arrive slightly faster or slower than the other frames. This is
called jitter, and degrades the perceived voice quality. To minimize this problem, configure the IP
telephone with a jitter buffer for arriving frames.
Note: You can change the jitter buffer on a configured IP telephone only if it is
connected to the BCM50, or if Keep DN Alive is enabled for an offline telephone.
This is how the jitter buffer works (assume a jitter buffer setting of five frames):
• The IP telephone firmware places the first five arriving frames in the jitter buffer.
• When frame six arrives, the IP telephone firmware places it in the buffer, and sends frame one
to the handset speaker.
• When frame seven arrives, the IP telephone buffers it, and sends frame two to the handset
speaker.
By using a jitter buffer, the arriving packets are delayed slightly in order to ensure a constant rate
of arriving frames at the handset speaker.
N0027269 01 N0027269 01