Abstract

Technological improvements have created a large need for increased optical backbone network transmission capacity. Higher launch powers into the fibre are one way to boost transmission throughput; however, this contributes nonlinear phase noise (NLPN) to the signal owing to fibre Kerr nonlinearity. This NLPN deteriorates the transmission performance, especially for higher order modulation and restricts data rates and transmission distance. To account for fibre nonlinearity, many digital signal processing (DSP) methods have been developed [1–4]. We demonstrate a reference constellation adaptation (RCA) algorithm for carrier phase recovery that can tolerate NLPN by simultaneously updating the filter weight w(n) for carrier phase tracking and reference constellation points d(n). Learning and adapting d(n) is performed to match them to the centroids of the distorted signal and thereby redefining the decision boundaries as shown in Fig. 1(a), where the black markers represent the adapted reference constellation points and black lines the corresponding decision boundaries. The block diagram of the proposed filter and the schematic for the simulation/experimental setup, along with the proposed DSP, are shown in Fig. 1(b) and (c), respectively.

© 2023 IEEE