Re: [10GBASE-T] The channel model
Chris,
For 100BASE-T and 1000BASE-T the channel bandwidth is 62.5MHz ,but the channel
is characterized up to 100MHz.This is done in order to cover all the
nonlinearities in signaling ,channel and to cover all the issues with egress
and ingress for EMI.
10GBASE-T proposed channel bandwidths are: 416MHz(PAM10) and 625MHz
(PAM5).Therefore in order to characterize the channel for all the none-ideal
effects(signal,balance,connectors,UTP,EMI etc) ,we need a guard band around
the signal bandwidth.There are already CAT6(CDT and others) cables today ,that
are characterized up to 550MHz.
Understanding the channel up to 1GHz will help with coding,scrambling and
defining the notch filter for EMI.
regards,pirooz
> Pirooz,
>
> Thanks for the input. I'm planning to schedule a con-call next week
> to discuss the channel modeling.
>
> What are the requirements that necessitate channel characterization
> up to 1 GHz?
>
> I'd like to limit the channel characterization to account for
> relevant signal spectrum and minimize excess bandwidth characterization.
>
> Regards,
>
> Chris
>
> ----- Original Message -----
> From: "Pirooz Hojabri" <pirooz@platolabs.com>
> To: <stds-802-3-10gbt@ieee.org>
> Sent: Tuesday, January 14, 2003 10:36 PM
> Subject: [10GBASE-T] The channel model
>
> >
> > Chris,
> >
> > I would like to propose the following steps in order to validate/generate
a
> > common channel model for 50m & 100m of CAT6/6e & CAT5e with two
connectors:
> >
> >
> > 1) Channel characterization for Attenuation, phase and roughness up to
1GHz.
> >
> > 2) Cable balance vs frequency up to 1GHz.
> >
> > 3) Channel characterization for NEXT,ANEXT,FEXT up to 1GHz.
> >
> > 4) Validate RLCG model based on above characterization.
> >
> >
> > regards,pirooz
> >
> >
> >
> >
> >
> > Plato Labs,Inc
> > Pirooz Hojabri
> > V.P Mixed Signal
> > 408-379-4900
> >
> >
Plato Labs,Inc
Pirooz Hojabri
V.P Mixed Signal
408-379-4900