Multilayer perceptron equalizer for optical communication systems

Optical fibers are now widely used in communication systems, mainly because they offer faster data transmission speeds, compared to any other type of digital communication system. Despite this great advantage, problems remain that prevent the full exploitation of optical connection: by increasing transmission rates over longer distances, the data is affected by nonlinear inter-symbol interference caused by dispersion phenomena in the fiber. Adaptive equalizers can be used to compensate for the effects caused by nonlinear channel responses, restoring the signal originally transmitted. The present work proposes a multilayer perceptron equalizer based on artificial neural networks. The technique was validated using a simulated optical channel, and was compared with other adaptive equalization techniques.