Functional link artificial neural network-based adaptive channel equalization of nonlinear channels with QAM signal

We investigate the application of artificial neural networks (ANNs) to adaptive channel equalization in a digital communication system with QAM signal. A novel computationally efficient functional link ANN (FLANN) is proposed for this purpose and its performance comparison with two other ANN structures (i.e., a multilayer perceptron and a polynomial perceptron network) along with a conventional linear equalizer trained with least mean squares algorithm has been carried out. The effect of the eigenvalue ratio (EVR) of the input correlation matrix on the equalizer performance has been studied. It is shown that the proposed equalizer structure outperforms the other two ANN structures and the linear equalizer in terms of the convergence rate, MSE floor and EER over a wide range of EVR and SNR conditions for both linear and nonlinear channel models