A New Blind Equalization Algorithm for M-PSK Constellations
Ali Ekşım, Serhat Gül · 2014
Abstract—In this paper, we propose an equalization algorithm for M-PSK constellations that greatly improves the convergence features and reduces steady-state error rate of the conventional Constant Modulus Algorithm (CMA). The proposed algorithm introduces a buffer and multiple step-size decision layers to the existing Variable Step Size Modified Constant Modulus Algorithm (VSS-MCMA) equalizer. The buffer is employed to combat the convergence issue while the additional layers have been introduced to improve sensitivity of the step size in both the convergence state and the steady-state. Computer simulations reveal that the proposed algorithm has better convergence rate than the VSS-MCMA and the CMA. Keywords—Blind equalization; constant modulus algorithm; step size I.