Re: [10GMMF] task 4 Oct. 4th call -- Polarization
I appreciate the work done by Joerg and his team, and the polarization
experiment report seems well organized, but...polarization penalty for
multimode links? I hope someone can educate me, because the way I see it,
polarization shouldn't make any difference to DMD.
I am starting with the assumption that only the input polarization was
changed, and everything else was unchanged -- i.e., the excitation region
didn't shift, the input field intensity pattern remained the same, etc.
DMD changes only if the power distribution among mode groups changes. A
mode group consists of one or more LP modes having nearly identical
propagation constant. My contention is that a change in input polarization
not only leaves power in each mode group intact, but it also leaves power
in each mode intact.
We normally characterize a mode as LP(m,n), where m is the azimuthal
number and n is the radial number. But actually, each LP(m,n) mode
consists of two (if m=0) or four (if m>0) distinguishable modes having the
same propagation constant -- distinguishable because of two possible
polarizations and two possible azimuthal orientations, sine and cosine.
For example, mode group 6 consists of LP(5,1), LP(3,2) and LP(1,3), and
has a total of 12 distinguishable modes.
Power in each LP mode is the sum of the powers carried by its
distinguishable modes. These distinguishable modes share the power, the
proportion being decided by how the input polarization aligns with the
polarization axes of the distinguishable modes. When the input
polarization direction is changed, the proportion of power among these
distinguishable modes changes, but the total power in an LP mode remains
the same. Therefore, DMD should not be affected by changes in input
Please tell me what I am missing. Thank you.
[mailto:owner-stds-802-3-10gmmf@IEEE.ORG]On Behalf Of yusun@IEEE.ORG
Sent: Wednesday, October 06, 2004 9:04 AM
Subject: [10GMMF] the minitues of task 4 Oct. 4th call
I attached the minutes of tele-conference of task 4 group on Oct. 4th. I
would appreciate your comments.
The proposed next task 4 meeting is
Oct 11th, 2004 (Monday), 11am east coast, 8 am west coast and 4 pm UK.