Adaptive zero-forcing blind equalisers
F.-B. Ueng · IEE Proceedings - Communications · 2001
Several adaptive blind equalisers, based on the zero-forcing (ZF) cost function, are presented. In order to estimate the mean square error (MSE) learning curve of a ZF FIR-type blind equaliser, a recursive method to predict the MSE trajectory is introduced. The ZF cost function on a decision-feedback (DF) structure is investigated and a convergence description for this DF structure is given. Finally, in order to speed-up the convergence of the proposed ZF blind equaliser, the multidimensional Newton search method is applied and the fast FIR and fast IIR type adaptive blind equalisers are derived. Some simulation examples are given to demonstrate the effectiveness of the proposed adaptive blind equalisers.