Neural detection of QAM signal with strongly nonlinear receiver

Neural receiver structures have been developed for adaptive discrete-signal detection in telecommunication applications. Neural networks combined with conventional equalizers improve the performance especially in compensating for nonlinear distortions. These distortions may result, for instance, fro m nonlinear amplification implemented for reducing the power consumption. I n this paper, the behavior of the neural receiver in multipath channel with additive white Gaussian noise has been investigated. The transmitted signa l is quadrature amplitude modulated (QAM). A receiver structure based on se lf-organizing map (SOM) is compared with a conventional decision feedback e qualizer (DFE). (C) 1998 Elsevier Science B.V. All rights reserved.