A novel decision feedback equalizer design based on generalized space translation

In this paper, a generalized space translation decision feedback equalizer (DFE) structure having arbitrary design parameters is presented. This new DFE structure extends previous research on design techniques based on space translation properties, such as the minimum bit error rate DFE (MBER-DFE) and the support vector machine DFE (SVM-DFE). A novel design technique, termed a Hybrid MMSE-SVM-DFE, based on generalized space translation is proposed. It combines the advantages of the minimum mean square error (MMSE) and SVM designs. Simulation results show that the proposed design performs very close to the MBER-DFE in terms of symbol error rate, over a wide range of signal-to-noise ratios.