Classification of Digital Modulation Using Wavelet Transform

Abstract

Many methods were presented for classification of digital modulation;these methods either deal with the received (modulated) or the basebandsignals. To reduce the classifier complexity, the first type is used. Discrete Fourier Transforms (DFT) or Fast Fourier Transforms (FFT) of the received signal may be used for such purpose. However, the drawback of this method is the time, which required for completing the operation, is proportional to the number of samples used for classification.In this work, a proposed classifier based on wavelet filters is used instead of the FFT classifier. Computer simulation tests show that the Haar wavelet filter is the most suitable type as compared with other types of wavelet filters. Extra tests show that the wavelet classifier has the same performance but it classify signals with bit rate higher than that of FFT classifier.