A neural network model applied to the detection of digital signals

This work presents a artificial neural network (ANN) approach for the signal decision problem at the receiver of digital communication systems with a modulation scheme whose signal elements belong to a finite bidimensional constellation. The decision system proposed, named Neural Decoder (ND), is a multilayer perceptron neural network trained with the backprogation algorithm and models a maximum-likelihood receiver. The ND training process as well as simulation results of their performance, regarding the conventional receiver, are presented for some modulation systems studied.