Neural Probabilistic Shaping: Joint Distribution Learning for Optical Fiber Communications

2025-07-21

AI Summary

Conventional probabilistic shaping (PS) for nonlinear optical fiber channels optimizes only marginal symbol distributions, leading to suboptimal capacity. To address this limitation, this work proposes an autoregressive end-to-end deep learning framework that jointly models and optimizes the *inter-symbol joint distribution*—the first such approach in optical communications. Trained and validated on dual-polarization 64-QAM over a single-span 205 km standard single-mode fiber link, the method achieves a 0.3 bit/2D gain in achievable information rate over state-of-the-art marginal-distribution PS. Its core innovation lies in embedding joint symbol distribution learning directly into the end-to-end communication system optimization pipeline, thereby overcoming the restrictive independence assumption inherent in conventional PS. This paradigm shift enables more efficient probabilistic shaping under strong nonlinear channel impairments, establishing a new foundation for capacity-approaching transmission design.

Technology Category

• Search and Optimization: Learning to Search
• Computer Vision: Learning & Optimization for CV
• Natural Language Processing: Learning & Optimization for NLP

Application Category

• User Modeling, Personalization and Recommendation: Psychology-informed user models and recommender systems
• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• Security and Privacy: Large-scale security measurements

Abstract

We present an autoregressive end-to-end learning approach for probabilistic shaping on nonlinear fiber channels. Our proposed scheme learns the joint symbol distribution and provides a 0.3-bits/2D achievable information rate gain over an optimized marginal distribution for dual-polarized 64-QAM transmission over a single-span 205 km link.

Problem

• Learning joint symbol distribution for fiber optics
• Improving information rate in 64-QAM transmission
• Optimizing nonlinear fiber channel performance

Innovation

• Autoregressive end-to-end learning approach
• Learns joint symbol distribution
• Achieves 0.3-bits/2D rate gain