Enhanced self-attention-assisted efficient multi-channel waveform modeling with stimulated Raman scattering effect.

Area of Science:

• Optical Communications
• Nonlinear Optics
• Signal Processing

Background:

• Stimulated Raman scattering (SRS) causes inter-channel power transfer and cross-channel coupling in optical systems.
• Kerr nonlinearities interact with power fluctuations, complicating C+L-band system modeling, especially with power pre-tilt.
• Accurate modeling of ultra-wideband (UWB) wavelength-division multiplexing (WDM) systems with nonlinear effects is challenging.

Purpose of the Study:

• To propose an efficient and accurate multi-channel waveform modeling method for WDM systems.
• To address challenges posed by SRS and non-flat launch power.
• To enhance generalization ability across various system conditions.

Main Methods:

• Developed an enhanced self-attention-assisted multi-channel waveform modeling approach.
• Incorporated rotary positional encoding into the attention mechanism's query (Q) and key (K) matrices to capture long-range dependencies.
• Utilized UWB WDM system simulations.

Main Results:

• The proposed method demonstrates strong generalization across different non-flat launch power profiles and transmission distances.
• In a 10-span link, the Q-factor deviation from the split-step Fourier method (SSFM) was only 0.31 dB.
• In a 5-span scenario, runtime was reduced by 99.4% with a Q-factor deviation of 0.24 dB from SSFM.

Conclusions:

• The enhanced self-attention model provides efficient and accurate waveform modeling for UWB WDM systems.
• The method effectively handles SRS and non-flat launch power, outperforming traditional methods in speed and accuracy.
• This approach offers a robust solution for complex optical communication system modeling.