Optical Fiber Communication Systems Based on End-to-End Deep Learning
This work addresses optical communication challenges for telecommunications, but it is incremental as it builds on existing deep learning methods with a novel application.
The authors tackled the problem of optical fiber communication over dispersive nonlinear channels by implementing an end-to-end deep learning system using bidirectional recurrent neural networks (BRNN) auto-encoders, achieving performance improvements as demonstrated in the first experimental demonstration.
We investigate end-to-end optimized optical transmission systems based on feedforward or bidirectional recurrent neural networks (BRNN) and deep learning. In particular, we report the first experimental demonstration of a BRNN auto-encoder, highlighting the performance improvement achieved with recurrent processing for communication over dispersive nonlinear channels.