Modulation Classification of MFSK Modulated Signals Using Wavelet Decomposition

Abstract

In this study,a wavelet decomposition based method is proposed for determining the modulation type of the incoming signal to the receiver which is one of the important problems in intelligent communication systems. In this method, it is aimed to design the transmitted signal for determining the type of Mary FSK modulated signal and to detect the energy in each frequency band by using Discrete Wavelet Transform (DWT). For this, standard deviations in the lower bands are as features. In order to evaluate the performance of the classifier, simulation studies have been performed at different signal-to-noise ratio (SNR) levels. When the results for different frequency settings, i.e. carrier frequency and frequency range, it is seen that the classifier using the K-means clustering algorithm has a higher correct classification performance than the results reported in the literature when the suitable carrier frequency and frequency range are selected.