Deep Learning Application for Classification of SEFDM Signals

Abstract

The paper considers the application of a convolutional neural network for the classification of SEFDM signals for different modulation schemes. A simulation model of the receiver and transmitter has been implemented for the case of a multipath channel and two frequency separations with steps of 0.1 and 0.2. For both cases, the classification accuracy values were obtained, which averaged 99% at signal-to-noise ratio equal to 10 dB.