Blind adaptation of decision feedback equalizers based on the constant modulus algorithm

Abstract

The constant modulus algorithm (CMA) is a simple and effective blind equalization algorithm that is generally implemented on feedforward linear adaptive equalizers. In this paper, we study the use of CMA on a decision feedback equalizer (CMA-DFE) for blind channel equalization. We analyze the convergence behavior of this highly nonlinear adaptive system. We also find a class of channels for which the decision directed DFE algorithm (DD-DFE) will not converge to the global optimum when initialized at the origin. For some of these channels, the CMA-DFE does not suffer the same ill-convergence.