Neural Probabilistic Shaping: Joint Distribution Learning for Optical Fiber Communications
By: Mohammad Taha Askari, Lutz Lampe, Amirhossein Ghazisaeidi
Potential Business Impact:
Makes internet signals travel faster and farther.
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.
Similar Papers
End-to-end Learning of Probabilistic and Geometric Constellation Shaping with Iterative Receivers
Information Theory
Makes internet signals send more data with fewer errors.
Probabilistic Shaping in MIMO: Going Beyond 1.53dB AWGN Gain With the Non-Linear Demapper
Information Theory
Boosts wireless signals, even with interference.
End-to-End Learning of Probabilistic Constellation Shaping through Importance Sampling
Information Theory
Makes internet faster by smarter data sending.