Add fiber with loss and CD only (disable SPM/XPM/FWM). Sweep launch power from -10 dBm to +10 dBm. The BER should improve with power (more OSNR) until you hit thermal noise. This curve is your linear baseline.
Simulate the Tx + Rx directly connected (0 km fiber). Adjust the LO power and receiver ADC bits until you hit the theoretical BER for your modulation format. If you can’t match theory here, you won’t match reality later. optiwave optisystem
The power emerges when you stop wiring components and start designing signal flow . Ask not "What block do I need?" but "How does the statistical distribution of my signal evolve?" Add fiber with loss and CD only (disable SPM/XPM/FWM)
Discipline is key. Start with datasheet values from real components (Finisar, II-VI, Broadcom). Add realistic connector loss (0.5 dB per mated pair). Include filter penalties from ROADMs. Add a safety margin of 2 dB OSNR to your target. This curve is your linear baseline