In this paper, we propose a novel blind equalizer which can deal with high-order modulated QAM (quadrature amplitude modulation) signals. This new scheme is based on the signal selection (SS) and the iterative weighted-least-mean-square (IWLMS) algorithm. The incurred additional complexity by the SS scheme is linear with respect to the sample size of the received signal and the IWLMS method is also very efficient. We employ numerous Monte Carlo experiments to compare our proposed blind equalization method with the popular constant modulus algorithm (CMA). The simulation results demonstrate that our proposed scheme is more robust than CMA especially when the QAM modulation order is large.
Kun Yan, Hsiao-Chun Wu, Dongxin Xu, S. S. Iyengar