On the Convergence of Blind Over-Sampled Decision Feedback Equalizers

Decision Feedback Equalizers (DFEs) efficiently reduce the effect of Inter Symbol Interference (ISI). Their blind (or Decision-Directed) adaptation may converge to so-called degenerated solutions when signal at the equalizer output is independent of its input. One of the solution to this problem is to use Predictive Decision Feedback Equalizer (PDFE), another solution proposed consists in imposing constrains on the coefficients of the feed-forward filter. These solutions are unsatisfactory so a new simple algorithm of blind adaptation is proposed. It consist in constraining the feedback coefficients and guarantees the convergence to desired solution. The convergence properties are shown theoretically and simulations results are presented.