An Accelerated Constant Modulus Algorithm

Abstract

— We present a novel adaptive algorithm for blind equalization. It is based on a tuner used in adaptive control that sets the second derivative of the parameter estimates and minimizes the cost function introduced by Go- dard. Based on simulation results we present a comparison with the Constant Modulus and the Shalvi-Weinstein algorithms. Both the convergence speed and the computational complexity of the proposed algorithm lie between those of the Constant Modulus and the Shalvi-Weinstein algorithms, thus present- ing a more favorable compromise between convergence speed and computational complexity. Some preliminary results also suggest that the proposed al- gorithm presents a more robust behavior with respect to convergence to global minima.