Geometric Constellation Shaping for Fiber Optic Communication Systems via End-to-end Learning
This work addresses the challenge of optimizing signal transmission in fiber optic systems for improved data rates and efficiency, representing an incremental advancement over existing geometrically shaped constellations.
The paper tackled the problem of geometric constellation shaping in fiber optic communication systems by developing an unsupervised machine learning method that embeds a differentiable fiber channel model within neural networks, resulting in improved performance up to 0.13 bit/4D in simulation and 0.12 bit/4D experimentally.
In this paper, an unsupervised machine learning method for geometric constellation shaping is investigated. By embedding a differentiable fiber channel model within two neural networks, the learning algorithm is optimizing for a geometric constellation shape. The learned constellations yield improved performance to state-of-the-art geometrically shaped constellations, and include an implicit trade-off between amplification noise and nonlinear effects. Further, the method allows joint optimization of system parameters, such as the optimal launch power, simultaneously with the constellation shape. An experimental demonstration validates the findings. Improved performances are reported, up to 0.13 bit/4D in simulation and experimentally up to 0.12 bit/4D.