That 24Kbps would make sense as G.729 should compress the voice to 8Kbps for a unidirectional voice signal, so 16Kbps for bi-directional.... 24Kbps should be plenty for overhead and a bi-directional voice stream. However, if you want you could use G.729b (or G.729ab if you happen to be using medium complexity) to enable Voice Activation Detection (VAD) and that can reduce your overall bandwidth utilization by up to 40% by not sending data when a party is not talking (and throw on some comfort noise while using VAD). That kicks azz, as that would lower the bandwidth needed for a G.729b (or ab) call to roughly 14.4Kbps, so you could (in theory) jam up to four simultaneous (close-to-toll-quality) calls over a single 64Kbps line. NICE!!
Mike W. "Steven A. Ridder" wrote in message [EMAIL PROTECTED]">news:[EMAIL PROTECTED]... > I'm not sure the context the document was written in, but it's only 24K > (give or take depending on the L2 encap) that you need to plan for. > > Steve > > > ""neil K."" wrote in message > [EMAIL PROTECTED]">news:[EMAIL PROTECTED]... > > Hi All, > > > > Using the standard formulas, I see for a bandwidth required for a g.729 > call > > is 24kbps without RTP compression.I used to do the same when using other > > codecs. > > Recently I came across some VoIP documentation which said that you will > have > > to consider two RTP flows to simulate a call, and hence the requirement > for > > Bandwidth doubles.say g729 24 kbps becomes almost 48kbps if u consider two > > RTP flows. > > > > Any help will be highly appreciated. > > > > Thanks, > > > > Neil Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=49388&t=49388 -------------------------------------------------- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
