Abstract:The first blind nonlinear equalizer using affinity propagation (AP) clustering is experimentally demonstrated for single-channel and WDM CO-OFDM. AP outperforms fuzzylogic c-means clustering and digital-back propagation for both QPSK and 16-QAM formats. IntroductionEndeavors to surpass the Kerr nonlinearity limit in long-haul coherent communications have been attempted in digital domain by Volterra-based nonlinear equalization (V-NLE) [1] and digital-back propagation (DBP) [2]. V-NLE and DBP however, can only tackle deterministic nonlinearities such as self-phase modulation, without considering the stochastic nonlinear interaction from polarization-mode dispersion and amplified spontaneous emission noise caused by cascaded optical amplifiers. On the other hand, full-step DBP (FS-DBP) is very complex and V-NLE shows marginal performance enhancement accompanied with a significant amount of floating-point operations, thus forbidding their implementation in real-time communications. Moreover, albeit the Kerr-induced nonlinear process is deterministic, in multicarrier schemes like coherent optical OFDM (CO-OFDM) the resulting nonlinear interaction between subcarriers becomes very complicated appearing random due to its high peak-toaverage power ratio (PAPR) [3]. Recently, unsupervised and supervised machine learning such as K-means clustering [4] and artificial neural network classification [3] have been introduced in optical communications to combat stochastic source of noises, performing blind and non-blind NLE, respectively. Here, we demonstrate the first blind-NLE using affinity propagation (AP) clustering for single-channel and WDM CO-OFDM. AP outperforms fuzzy-logic C-means (FL) and K-means clustering, as well as digital deterministic solutions such as FS-DBP and V-NLE, by reducing a significant amount of stochastic nonlinear noise on middle subcarriers.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.