Contents I. Introduction
Electrical equalization techniques are good choices for optical fiber communication (OFC) systems because of their potentially cost-effective realization in receivers. Recently, iterative (turbo) equalization and maximum-likelihood sequence estimation (MLSE) have been discussed intensively, since they perform effectively in mitigating the intersymbol interference (ISI) caused by polarization-mode dispersion (PMD) and chromatic dispersion (CD) [1]–[4]. Iterative equalization is an approach to exchange soft information between the equalizer and decoder iteratively to realize the joint work of the two sections. Iterative equalization algorithm based on the minimum mean square error (MMSE) criterion, called iterative minimum mean square error (IMMSE) equalization, was first proposed by Tüchler et al. to mitigate ISI in linear additive white Gaussian noise channels [5]. This algorithm has outstanding performance but has not been available for OFC systems yet, since the photodetection at the receiver transforms linear distortions in the optical fiber channel to be nonlinear and leads to the non-Gaussian noise property. In this letter, accounting for nonlinearity as well as non-Gaussian noise, IMMSE equalization for OFC systems is derived. Numerical simulations are carried out to demonstrate the effectiveness of the proposed algorithm. Compared with MLSE equalization, IMMSE equalization presents a better capability of mitigating the ISI in optical fiber channels.
Low-pass equivalent model of the optical fiber transmission system.
