Piecewise Feedforward Neural Network Based Nonlinear Equalizer for Short-Reach DML-DD System

Abstract

We propose a targeted piecewise feedforward neural network (PW-FNN) to mitigate both the linear and nonlinear distortions in intensity modulation direct detection (IM-DD) system based on direct modulated laser (DML). By this equalizer, a 56Gbit/s pulse amplitude modulation (PAM)-4 transmission over 30km standard single-mode fiber (SSMF) is experimentally demonstrated with a 16.8GHz DML with bit error rate (BER) below 7% hard-decision forward error correction (HD-FEC) threshold of 3.8 × 10−3. Compared with Volttera and traditional FNN equalizer, the proposed PW-FNN equalizer exhibits more than 2dB receiver sensitivity improvement. The experimental results validate that the proposed PW-FNN can be a promising solution for short reach DML-based IM-DD system.