Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: [10GBT] Channel models


Your statement is  partially correct as long as it refers to FEXT and
not to ELFEXT, opposite to the statement made earlier in the comment
Valery Rybinski. In fact FEXT is defined as input-to-output far-end
crosstalk.  As such it is reciprocal if the ends are reversed, i.e. if
the sending and receiving end are changed on the same pairs.  In the
cable industry we use the terms "IFEXT" and "OFEXT" to denominate the
inside end and the outside end of the cable on the drum.  Thus IFEXT ij
is equal to OFEXT ji but not as Valery stated symmetrical at the same
end.  This is simply a result of the coupling coefficients which vary
along the cable, partially stochastically partially deterministically.
 The propagation constants go into the crosstalk equations as well.  As
to the reflection problem is at least for the crosstalk interesting, as
it implies for short cables between connectors multiple reflections.
 Here we have again, and very urgently the problem of forward echo and
the corresponding reflection coefficients, though in both directions. I
am working on a complete model for the channel, but this may still take
some time before completion.

As to the common mode terminations commonly used, they affect also
slightly the crosstalk, but much more the balance and the thereout
resulting conversion losses due to mode conversion.  Though these
problems may be resolved by simulation, the hick-up remains, that you
will not be able to repetitively and correctly measure common mode

I attached as well a short comment regarding the definition of ELFEXT
which is in the section 55 defined incorrectly.  I mention also at the
same time the problem of "insertion loss" versus "operational
attenuation" and attenuation.  The term "insertion loss" should be
avoided and replace by operational attenuation, as it is not a value
which can be simply added, as unfortunately and misleadingly cited and
printed (paper is patient!) in different standards.

Kind regards,

Jo Walling

CDimi80749@AOL.COM wrote:

> Sanjay,
> Short answer:
> The traditional simple transmission line models utilized to predict
> far-end crosstalk (below 100 MHz) do
> not account for common mode propagation (and termination) and neglect
> far-end reflections. Many of the
> traditional references assume that like pairs in a cable have
> identical propagation constants (magnitude and phase),
> and the lines are terminated in their characteristic impedance.  These
> effects (combined) can result in measurable
> differences between the FEXT of pair ij and the FEXT of pair ji.
> Regrads,
> Chris
> In a message dated 5/12/2004 6:40:31 PM Eastern Standard Time,
> SKasturia@TERANETICS.COM writes:
>     Bryan,
>     Good question. I assumed that FEXT between pair i and j would be the
>     same as between j and i. Thinking about it some more, I am not sure if
>     that is correct. Can others let us know what measurements have
>     revealed?
>     Chris DiMinico, can you answer this?
>     Sanjay
>     cell (650) 704-7686
>     office (408) 653-2235
>     -----Original Message-----
>     From: stds-802-3-10gbt@IEEE.ORG [mailto:stds-802-3-10gbt@IEEE.ORG] On
>     Behalf Of Sparrowhawk, Bryan
>     Sent: Wednesday, May 12, 2004 4:19 PM
>     To:
>     Subject: Re: [10GBT] Channel models
>     Sanjay,
>     I have a question about your terminology regarding far end
>     crosstalk in
>     "c)".
>     Could you have meant twelve pair combinations for FEXT instead of 6?
>     Or do you speak of "Far end crosstalk" as others speak of 'reverse
>     NEXT'?
>     Bryan S.
>     -----Original Message-----
>     From: Sanjay Kasturia [mailto:SKasturia@TERANETICS.COM]
>     Sent: Friday, April 09, 2004 3:36 PM
>     To:
>     Subject: [10GBT] Channel models
>     We have four channel models defined from the March 2004 plenary
>     meeting.
>     For detailed PHY performance simulations, we will need sample channel
>     data that complies with these models.
>     We will greatly appreciate it if any task force participants can
>     provide
>     full matrix (4x4) channel data. This would ideally include:
>     a) Insertion loss data on each of four pairs - real and imaginary
>     b) Near end crosstalk coupling (6 unique pair-combinations) - real and
>     imaginary
>     c) Far end crosstalk coupling (6 unique pair-combinations) - real and
>     imaginary
>     d) Four unique echo transfer functions (one for each pair) - real and
>     imaginary
>     This can be obtained by measurements on actual cable and then
>     scaling to
>     meet the specified channel models.
>     Any cable companies volunteering to provide this data?
>     There is some data from the cabling Ad-hoc available on:
>     This data has some of the items listed above but not all. It does,
>     however, provide enough to get some sense of performance.
>     Sanjay Kasturia
>     cell (650) 704-7686
>     office (408) 653-2235

FEXT - Comments.doc