Machine learning for short reach optical fiber systems

Short reach optical fiber communications rely on the intensity modulation/direct detection (IM/DD) technology to enable simple and inexpensive solutions in many data center, metro and access network scenarios. The IM/DD links are severely impaired by fiber dispersion as well as non-linearity and noise stemming from the low-cost transmitter and receiver hardware components. Thus, digital signal processing (DSP) is required for increasing the data rate and transmission reach of such systems. However, there is a lack of optimal, computationally feasible DSP algorithms for communication over the dispersive non-linear IM/DD channel. This necessitates the use of carefully chosen approximations to better exploit the data transmission potential over the optical links. The chapter discusses the application of artificial neural networks (ANN) and deep learning in the design and optimization of DSP modules for short reach optical IM/DD communication systems. In addition to the design of a deep learning-based receiver for conventional pulse amplitude modulation (PAM) transmission, the focus is on the implementation of fully learnable ANN transceivers, also known as auto-encoders.