Skip to Content.
Sympa Menu

wg-voip - Re: Help needed on VoIP project

List archive

Re: Help needed on VoIP project


Chronological Thread 
  • From: Guy T Almes <>
  • To: Walt Magnussen <>,
  • Cc: ,
  • Subject: Re: Help needed on VoIP project
  • Date: Mon, 01 Mar 2004 09:58:00 -0500

Walt,
Excellent.
-- Guy

--On Monday, March 01, 2004 08:49:10 -0600 Walt Magnussen <> wrote:

Jeremy, Guy, et.al

The network requirements testing is one of the tasks that we are
working on as we speak. We have purchased an Agilent Voice Quality
Tester (VQT) and are creating a testbed that will measure the impact of
Jitter, Packet Loss and Latency on Voice Quality (the pudding test).

We will keep you posted on progress on this one.

Walt


Walt Magnussen, Ph.D.
Director for Telecommunications
Texas A&M University

979-845-5588 phone
979-777-0999 cell
979-847-1111 fax

Jeremy George
<>
03/01/04 08:40AM >>>

Turning over the underlying network requirements would be a huge
help for those of us contemplating running VoIP. It would be great
if the ITEC really looked into it.

There was an interesting email discussion around last October in the
ietf sipping group on draft-ietf-ieprep-framework-06.txt in re jitter
in
VoIP. The conclusion, if such a term can be ascribed to anything in
an ietf mailing list :), was that jitter, per se, was not a critical
issue. Delay variation needs to be understood in relation to jitter
buffer depth. It's the depth of the buffer that's critical. Too
little and you get snap, crackle and pop. Too much and you introduce
unacceptable delay.

Also, "packet loss" in VoIP is a textured term that needs to be
understood in the context of maximum acceptable M-to-E delay. In
many contexts a packet is "lost" (e.g. undeliverable) after 150ms.

- Jeremy


On Mon, 1 Mar 2004, Guy T Almes wrote:

Date: Mon, 01 Mar 2004 09:15:22 -0500
From: Guy T Almes
<>
To: Jon Zeeff
<>,
Walt Magnussen
<>,

Subject: Re: Help needed on VoIP project

Jon,
I agree.
Robustness of user experience in the presence of
good-but-perhaps-not-very-good IP service is one way to put it.
Regards,
-- Guy

--On Monday, March 01, 2004 07:54:31 -0500 Jon Zeeff
<>
wrote:

>
> This is somewhat off topic, but to me, the most important thing to
test
> is resistance to packet loss and jitter.
>
>> VoIP project. One of the 2004 projects is to evaluate SIP
>> interoperability between various vendors SIP based instruments and
other
>> vendors SIP based servers. The goal is to document the level of
>> interoperability. Listed below is a partial list of desired
voice
>
>



--







Archive powered by MHonArc 2.6.16.

Top of Page