Self-learning deconvolution using a cascade of magnitude and phase equalizers

In this work, we propose a non-linear structure for self-learning equalization, which can be easily updated using the direct-decision error criterion. Such a structure consists of three different systems: an IIR predictor that provides the magnitude equalization, an automatic gain control and a non-linear phase equalizer. The paper presents a theoretical analysis for the proposed structure and some simulation results with severe channels