Re: [10GMMF] Re feedback Polarisation effect and my HMG calculations
Dear Yu,
I agree on your comments. I have just replied to John on the same way
you are justifying polarization dependent pulse response. I guess Joerg
simulation he did yesterday highlights this approach.
Essentially, higher order modes reaching outer core region where
multiple alpha or ripple are present become no more "groupable" due to
their different propagation constant. If polarization affects individual
modes this at least gives clearer indication to our picture
(hopefully...)
Best regards
Stefano
-----Original Message-----
From: owner-stds-802-3-10gmmf@IEEE.ORG
[mailto:owner-stds-802-3-10gmmf@IEEE.ORG] On Behalf Of Yu Sun
Sent: Donnerstag, 4. November 2004 21:28
To: STDS-802-3-10GMMF@listserv.ieee.org
Subject: Re: [10GMMF] Re feedback Polarisation effect and my HMG
calculations
Dear John,
Thank you very much for your comments.
For a perfect radial symmetric waveguide, the change of input
polarization varies the energy of the degenerate modes (the sin mode and
cos mode) in one mode. The total power of the mode is a constant. The
root of the polarization effect is the imperfection of the fibers, the
asymmetric fiber core, the perturbation of the index profile, as well as
the external stress. These imperfections introduce the modal selective
loss, mode coupling and power diffusion in the multimode fibers. The
exact mathematic expression of these behaviors is complicated. However,
the common point of these behaviors is that the energy in the modes is
redistributed and how the energy is redistributed depends on the spatial
distribution of the modes and their initial power partition.
Empirically, the combination of polarization induced energy transfer
among the degenerate modes and modal selective lose, such as an offset
at the connector or detector, resembles the experimental observation.
I hope this make some sense. I would appreciate your comments.
Best regards,
Yu
-----Original Message-----
From: owner-stds-802-3-10gmmf@IEEE.ORG
[mailto:owner-stds-802-3-10gmmf@IEEE.ORG] On Behalf Of Abbott, John S Dr
Sent: Thursday, November 04, 2004 1:40 PM
To: STDS-802-3-10GMMF@listserv.ieee.org
Subject: Re: [10GMMF] Re feedback Polarisation effect and my HMG
calculations
David, thanks for your kind note.
I would comment to the overall group that the mathematics gets
burdensome in some of this and David's care is appreciated. For
example, his equation
(27) corrects a typographical error in the Saijonmaa reference.
I think that maybe in the observation of polarization sensitivity what
is happening is that one of our assumptions about the laser launch &
light propagation in FDDI links is not a sufficiently accurate
approximation
These are some of the ideas which have occurred to me:
A. is the launch axis parallel to the fiber axis? I think the Grau
paper in David's reference list suggests that if the launch has some
tilt that there will be a polarization effect, and one explanation is
that the launch is extremely sensitive to tilt (i.e. we can never
eliminate tilt) so that we cannot eliminate some residual polarization
sensitivity. B. similarly, David's revival of the Hermite-Gaussian modes
suggests the idea that a very small core ovality might break the
symmetry presumed in much of our analysis, and that as soon as that
happens polarization effects can be important. Since worst-case FDDI
fibers can have a core ovality of 6%, one could definitely imagine this
effect; however, Agilent and Infineon seem to see the effect on every
fiber with a perturbation which they look at.... C. There is a dn/dr
polarization correction term (the scalar wave equation ignores an extra
term in the vector Maxwell's equations) which can help explain a
polarization effect in fibers with large perturbations near the center;
however, it sounds like Agilent & Infineon see effects with an offset
launch if there is an index perturbation at that radial position.
Of these, it sounds like quantifying the sensitivity to ovality might be
the most useful, but maybe it is worthwhile brainstorming other ideas.
John A.
-----Original Message-----
From: david_cunningham@agilent.com [mailto:david_cunningham@agilent.com]
Sent: Wednesday, November 03, 2004 11:41 AM
To: AbbottJS@corning.com; STDS-802-3-10GMMF@LISTSERV.IEEE.ORG
Subject: Re feedback Polarisation effect and my HMG calculations
John,
Per your feedback I have converted the calculations of my earlier
discussion document called "Variation of the power coupled to the mode
groups of a circular core square law multimode fibre from a circular
single spatial mode laser" from HMG into LGG format (see attachment for
amended document).
The results now agree with your feedback. Specifically:
1) The excited mode power distribution (MPD) is constant and independent
of the orientation of the optical polarisation of the laser source.
2) The coupled power is not equi-partitioned between the individual
modes within a group.
3) At the launch, the optical polarisation of the excited fibre modes
are the same as that of the exciting laser.
4) For axial-symmetric refractive index distributions the impulse
response remains constant when the angle of optical polarization is
rotated.
This is because the modes within each group which exchange power when
the polarisation is varied are the cosine and sine modes. As you have
already said for axisymmetric index perturbations these mode have the
same delay time.
I am now right back where I was about a month ago - I don't see how
axisymmetric models can be used to calculate the change in impulse due
to launch polarisation variation.
Thanks for the feedback,
David