Deep learning-based nonlinear phase shift estimation in coherent optical communication systems

Abstract In this paper, a convolutional neural network (CNN) based algorithm is proposed for the estimation of nonlinear phase shift (NLPS) in coherent optical communication systems. Simulations show that the mean relative errors (MREs) of the estimated NLPS will be less than 5.0%, 2.0% and 1.0% as the optical signal-to-noise ratio (OSNR) is larger than 18 dB, 22 dB and 27 dB, respectively. Furthermore, we numerically investigate the NLPS estimation scheme for signals with data rates of 56Gbit/s, 70Gbit/s and 112Gbit/s respectively, and the MRE analysis shows that the proposed algorithm is insensitive to the variation of data rates in the PDM-QPSK systems.