300 meters on installed MMF
It seems that Part IV of the presentation ("Coding Gain")
is corrupted (I downloaded it but I cannot read it). I sent a
replacement to David Law a few minutes ago.
Part IV contains the results of long link simulations of
the PAM-5+4-WDM at 1.25 Gbaud/s system run at
various SNRs with the corresponding measured BERs,
in an attempt to measure the actual coding gain
of the present implementation (and compare it with
the ideal coding gain of 6 dB).
The accompanying graphs in this Part IV demonstrate
how robust and powerful the Trellis Coded Modulation
technique is: the Viterbi decoder recovers the data
when the eye opening at the receiver is practically
Jaime E. Kardontchik
San Jose, CA 95131