Modified Iterative Decision Feedback Equalization for Communication Systems

A number of reduced-complexity methods for turbo equalization have recently been introduced in which MAP equalization is replaced with sub optimal, low complexity approaches. In this paper, we investigate the inefficiency of the conventional decision feedback algorithm when higher-level modulation is used in Inter Symbol Interference channels. We estimates the data using the apriori information from the SISO channel decoder and also the a priori detected data from previous iteration to minimize error propagation. The proposed algorithm for 8-phase shift keying, 16-phase shift keying and 64 quadrature amplitude modulation was verified by assuming a statistical model for the a priori information combined with equalizer output to improve the interference cancellation. The simulation results show that the proposed low complexity DFE algorithm improves considerable compared to the conventional DFE when higher-level modulation are used. It was shown that at higher level modulation, the modified DEF has 1.6dB gain over the conventional DEF after seven iterations.