Blind Equalization with NARX Neural Networks

This paper presents a study on the blind equalization of Bipolar 2-PAM and 4-QAM signals using NonLinear Autoregressive Exogenous Input (NARX) neural system. The paper can be separated into two sections, the initial segment includes the theoretical issues about the blind equalization and neural equalizers, the subsequent part addresses the problem of software implementation. Also, the paper proposes a blind equalization model that combines the concept of decision-feedback equalizer and the NARX neural networks. Simulations with 2-PAM and 4-QAM signals have been carried out to test the performances of the proposed model.