Channel Estimation Methods Based on Volterra Kernels for MLSD in Optical Communication Systems

Abstract

Maximum likelihood sequence detection (MLSD) is the most effective electrical domain equalization scheme for mitigating dispersive optical channel impairments such as chromatic dispersion or polarization-mode dispersion. Parameter estimation for MLSD is not straightforward in optical communication systems due to the square-law nature of photodiodes. We propose a simple and efficient channel parameter estimation scheme for MLSD based on Volterra kernel modeling of the nonlinear distortion of the electrical postdetection signals.