Min-norm based alphabet-matching algorithm for adaptive blind equalisation of high-order QAM signals

International audience; An adaptive blind equalization algorithm for high-order quadrature amplitude modulations (QAM) is proposed. It is derived by combining the multi-modulus criterion and a new alphabet-matching penalty term. This penalty is based on the minimum L1-norm of the deviations of the equalizer output from the alphabet elements. Numerical experiments show that the proposed criterion provides higher performance in terms of residual error and convergence speed than the conventional multi-modulus algorithm (MMA) and its penalized versions.