Joint Tracking of Polarization State and Phase Noise Using Adaptive Cascaded Kalman Filtering

Abstract

We propose a cascaded Kalman filtering (CKF) technique, a series of extended Kalman filter (EKF), and a linear Kalman filter, for joint tracking of phase and amplitude noise as well as polarization state. We further propose an adaptive CKF (A-CKF), in which the process noise covariance has been estimated adaptively, using covariance matching method. Numerical results on back-to-back and transmission scenarios on 28-Gbd polarization-multiplexed 16-quadrature amplitude modulation system justify that our proposed CKF as well as A-CKF outperforms the earlier proposed EKF and the unscented Kalman filter (UKF) with an improved tolerance toward polarization rotations and phase noise besides less computational effort.