Joint Carrier-Phase Estimation for Digital Subcarrier Multiplexing Systems With Symbol-Rate Optimization

Digital subcarrier multiplexing (SCM) has recently emerged as a promising solution for next-generation ultra-high-baudrate coherent optical communication systems. Among its distinctive advantages over traditional single-carrier modulation, SCM enables the exploitation of symbol-rate optimization (SRO), which has been shown to enable the passive mitigation of the nonlinear interference noise (NLIN) that is generated during propagation over dispersion-unmanaged optical fiber systems. However, the full exploitation of SRO-based NLIN mitigation is severely hindered by the uncompensated distortion caused by laser phase noise (LPN) and non-linear phase noise (NLPN), whose impact is magnified by the use of low-baudrate subcarriers. Resorting to low-complexity carrier phase estimation (CPE) algorithms, in this paper we experimentally demonstrate that it is possible to overcome the hurdles posed by LPN and NLPN in SCM systems, provided that adequate joint-subcarrier CPE processing is employed. A dual-stage joint-processing approach composed of a pilot-based CPE optionally followed by a blind phase search (BPS)-based estimator is implemented and experimentally assessed, enabling to effectively optimize the symbol-rate per subcarrier down to 3 GBaud, in accordance with the theoretical SRO predictions for the system under test. In addition, we demonstrate that signal-to-noise ratio (SNR) gains of more than 1 dB can be achieved through joint-subcarrier CPE processing in shorter-reach links, while this gain tends to progressively reduce with increasing propagation distance, down to about 0.5 dB gain after 3000 km propagation.