Neural Network Based Receiver Design for Software Defined Radio over Unknown Channels

Abstract

In communication systems, the channel noise is assumed to be white and Gaussian distributed. Therefore, in general practical systems, optimum receiver structure designed for the additive white Gaussian noise (AWGN) channel is employed. However, in wireless communication systems, noise is often caused by a strong interferer, which is colored in nature. Color of the noise is defined as the variation in power spectral density in the frequency domain. Designing the optimum receiver for different channel models is difficult and not reasonable because channel model is not known at the receiver and channel statistics are needed. In this paper, we propose neural network (NN) based approach to demodulate the transmitted signal over unknown channels. Simulation results in various signal environments are presented to the performance of the proposed system. It is shown that the proposed approach has the same performance with the conventional demodulator structure for AWGN channels while it has clear advantage for unknown channel models.