About | Lab | Neigborhood | Most popular | Help us | Forums

Panoramisk / The VoIP druid 

Annonce

On peut être passionné par la téléphonie et prendre soin de soi : "Faites du sport, faites du vélo" est la devise du site d'e-commerce lancé par l'un des auteurs de Panoramisk : bikeo. Si vous faites du vélo, que ce soit en ville, sur la route ou sur les chemins plus accidentés, n'hésitez pas à aller faire un petit tour sur www.bikeo.fr pour vos prochains achats verts.
En plus leur plate-forme de téléphonie utilise Asterisk, comme quoi on peut la convergence est une réalité.

Voice over IP and Wi-Fi?

Why not, Wi-Fi can transport data frames, but not efficient.

Voice is by nature an analog signal that requires to be digitalized to be transported on modern digital networks. This process is available for long now and implemented in specific components called codecs. These codecs are able to build small chunk of digital voice corresponding to samples of about 20ms. When transporting voice over IP networks, the principle is to encapsulate the sample out of the codec into an RTP frame, RTP is using UDP as transport mechanism over IP. The overall encapsulation size in order to transport 20ms of voice is expensive: over an Ethernet network a standard voice call using a G.711 codec (64Kbps PCM modulation) is using roughly 80Kbps of bandwidth.

On a Wi-Fi network, the encapsulation is the same and the consumption is the same as on Ethernet networks, but each frame before being send is following a complex path called the media access. The main rule over the Wi-Fi network is that each station should wait the media (radio) to be free and then wait its turn to be able to talk. The delay before sending a frame in the best case is 50µs. Furthermore, each date frame should be acknowledged over a Wi-Fi network. This means that you cannot send your next frame unless the previous acknowledge came back, in the case of a voice traffic requiring a flow of 50 frames per second this mechanism wastes a lot of time.

The last main issue relates to mobility. On a network composed of multiple access point (an ESS) each client can change its association towards an access point proposing a better signal quality. This roaming takes some time that can go from tenth of a second to some seconds. When moving with a laptop from a cubicle to a meeting room, this delay is not sensible, but when talking over the phone, you will hear a glitch. If you compare to a DECT or GSM solution, Wi-Fi proposes a lower quality with regards to roaming.

Wi-Fi has not been imagined to transport voice traffic, but the broadband throughput proposed by Wi-Fi networks, even at 11Mbps is far enough to transport a voice phone call, furthermore, adaptative codecs are available to handle frame loss, jitter and delays liked to mobility.

Posted by: Alexandre Chauvin-Hameau, on 12/05/2006
Trackback | Popularity: 10%
tagged , , and
AddThis Social Bookmark Button
UselessNothing newInformativeLearned a lotAmazingly helpful
Loading ... Loading ...

See also

And why not

Leave a comment

© 2008 Panoramisk | Creative Commons License wordpress logo