A robust and low complexity clustering-based blind equalizer for PolMux QAM optical coherent systems

In polarization-multiplexed (PolMux) coherent systems, adaptive blind equalizer is efficient in demultiplexing and mitigating inter-symbol interference. A novel clustering-based blind algorithm is proposed, which has simpler computational requirement than the traditional clustering-based blind algorithms. And furthermore, the variance of Gaussian clusters and step size in the novel algorithm can adjust adaptively in the process of convergence, which means it has excellent robust ability. By simulation in a PolMux-16/64QAM coherent systems, we show that the proposed algorithm outperforms existing blind algorithms.