Re: [ns] Model for voice-over-ip

[Carlos Alberto Kamienski <cak@cin.ufpe.br>]

Hi George

It's based on an ITU-T recommendation (P-59)
Have a look at
http://www.cs.berkeley.edu/~chuah/research/phd.html#clearinghouse

references:
 QoS Provisioning Using A Clearing House  Architecture
and
Providing End-to-end QoS for IP-based Latency-sensitive Applications

The data rate during On periods depends on the codec you are using and on protocols
you are using (because you have to take into account protocol headers).
I used those parameters I presented you and data rates of 80 Kbps for On periods
and 0 Kbps for Off periods.

Regards

Carlos

George Riley wrote:

> Hi Carlos, thanks for the advise.  Two questions..where did you get this
> info (is it published somewhere)? and what is the data rate during the on
> period (in other words, how many bytes/sec should be generated during the
> on period)?
> Thanks.
> George
>
>
>                         -George F. Riley (riley@cc.gatech.edu)
>
> On Tue, 19 Jun 2001, Carlos Alberto Kamienski wrote:
>
> > Maybe you could use On-Off sources (Application/Traffic/Exponential)
> > with On periods (burst-time) of 1.004 s and Off periods (idle-time) of
> > 1.587 s.
> >
> > Carlos
> >
> > On Tue, 19 Jun 2001, Debojyoti Dutta wrote:
> >
> > > You could use a real audio model. It could be used as an approximation.
> > > Isnt it ?
> > >
> > > Debo
> > >
> > > On Tue, 19 Jun 2001, George Riley wrote:
> > >
> > > > Does anyone have (or know of) an ns model for voice-over-ip traffic?
> > > > Suppose I want to model 60 seconds of voice traffic..what i would like is
> > > > a "voice application" that sent bursts of data that (more or less) match
> > > > what one would expect for a VOIP end system.
> > > > Ideas?
> > > > George
> > > >
> > > >
> > > >                   -George F. Riley (riley@cc.gatech.edu)
> > > >
> > > >
> > >
> > >
> >
> >