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

[10GMMF] TP2 penalty analyses

Per the last TP2 con-call, I'm trying to do penalty analyses on a variety of candidate TP2 waveforms. Here are my plans.
  • How do various TP2 waveforms interact with (Cambridge) fibers to cause penalties in an EDC system? There has been some work in this area, but more data is needed.
  • At some point, we will have decide where to draw the spec (budget) limit at 80% cumulative coverage.
  • What speed of Gaussian  matches the budget limit (for TP3 test)?
  • Can TP2 and fiber penalties be separated as independent measurements and budget terms?
Variables, inputs
  • Variety of TP2 waveforms with relevant ranges of parameters
    • PRBS7-10, at least 8-16 samples per UI
    • Simulated Gaussian with varied rise times
    • Simulated rate equation waveforms
      • variety of operating conditions
      • also with electrical mismatches and losses
    • Experimental trace captures with real lasers, averaged >=16
      • variety of operating conditions
      • also with drivers, and coupling circuits
  • 108 fibers, 3 offsets - 324 cumulative cases
  • 220 m, 300 m
  • # of FFE and DFE taps (question for the committee - we could run both for now)
    • Large #'s approximate PIE-D
    • or should we set something realizable like 10/4?
Data out
  • Plots of cumulative coverage of 324 cases vs. penalty
  • TP2-only penalty + fiber-only penalty vs. total penalty, at 80% coverage (to test the question of independence)
To proceed, I encourage any comments and suggestions on this outline. ClariPhy is generating a range of TP2 waveforms, but I also welcome other waveforms as described above, or any additional useful variations.
Tom Lindsay
ClariPhy Communications
phone: (425) 775-7013
cell: (206) 790-3240