A decision feedback equalizer utilizing higher-order correlation

A decision feedback equalizer (DFE), which utilizes higher-order correlation (HOC) information available in the input vector of the DFE is proposed. Due to its better utilization of the available information, better performance is expected and then demonstrated from several respects. The capability of the proposed structure in forming nonlinear decision boundary is shown. Better mean-square error (MSE) performance is presented. Better bit-error rate (BER) performance is demonstrated. Less performance loss due to error propagation and less effect caused by improper parameter setting are observed.<>