Blind Neural Network Equalizer Based on QAM and Constant Modulus Algorithm

By using QAM signals as input, this paper adopts a blind equalizer based on neural network and constant modulus algorithm. By very few training serial signals to make the network convergent, and then the equalizer changes to the blind algorithm. The simulations show that this equalizer has better performance whether at convergence speed or the remnant errors´ energy, and its convergence capability is steady.