Beyond 200 Gbit/s/λ VSB PS-PAM8 Employing Joint Neural Network Equalization at C-band

Abstract

We experimentally demonstrate a high-speed vestigial sideband (VSB) probability shaping 8-ary pulse amplitude modulation (PS-PAM8) system at C-Band. The strong-coupled linear and nonlinear impairments in high-speed VSB transmission systems lead to performance degradation. Neural network-based equalization has been widely adopted in the optical transmission system. However, the effective combination of linear and nonlinear impairment compensation in NN equalization has not been discussed intensively. In this experiment, we have proposed an equalizer based on the joint neural network (JNN) to compensate for the linear and nonlinear impairments. With the proposed JNN equalizer and PS technique, 92 GBaud PS-PAM8 signals can be delivered over 10 km standard single mode fiber (SSMF). Given the forward error correction (FEC) overhead and the entropy of PS-PAM8 signals, the net bit rate can reach 207 Gbit/s/ <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\lambda $ </tex-math></inline-formula> , satisfying the demand for a 200 Gbit/s/ <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\lambda $ </tex-math></inline-formula> data connection.