An intelligent classification system based on wavelet transform for digital modulations

The digital modulation recognition is an important topic for communication system. In this thesis, the digital modulation applications, which are conducted by using optimum wavelet entropy parameter values are presented. A genetic- wavelet ?neural network(GWNN) model is developed in here. GWNN includes three layers which are genetic algorithm, wavelet and multi-layer perception. The genetic layer is used for selecting the feature extraction method and obtaining the optimum wavelet entropy parameter values. The wavelet transform layer consists of two part: wavelet decomposition and wavelet entropies. The multi- layer perceptron layer is used for evaluating the fitness function of the genetic algorithm and for classification digital modulation. Keywords: Modulation recognition, adaptive feature extraction, wavelet decomposition, entropy, genetic algorithm, artificial neural network, expert system