We are pleased to announce the publication of our recent paper:
S. Srivallapanondh, P. Freire, G. Parisi, M. Devigili, N. Costa, B. Spinnler, A. Napoli, J. E. Prilepsky & S. K. Turitsyn,
“Low complexity neural network equalizer for nonlinear mitigation in digital subcarrier multiplexing systems,” Optics Express, Vol. 33, No. 2 (2025).
🔗 Access the article: https://doi.org/10.1364/OE.542061
Highlights:
- Introduces a combination of bi-directional LSTM and 1D CNN layers for equalization in DSCM systems.
- Applies a weight-clustering compression method to drastically reduce complexity (up to ~91%) while preserving performance.
- Provides a promising path toward real-time, efficient neural network equalizers in high-speed optical transport systems.