M-ary Quadrature Ampplitude Modulation Classification Using Skewness and Kurtosis

The aim of modulation classification is to identify the modulation scheme of unknown modulated signals. In this paper, we present a new modulation classification approach which has two main steps. First, the valuable information of the signals is extracted by utilizing skewness and kurtosis as a features extraction. The extracted features are then introduced to a hybrid system which combines a neural network and a fuzzy inference system to identify the modulated signal. The proposed M-QAM signals classification system was investigated that identifies 16QAM, 32QAM, 64QAM, 128QAM, 256QAM, in presence of white Gaussian noise channel at -2, 0, 2, 4, 6, 8, 10 dB signal to noise ratio. MATLAB programs were designed to fulfill all the tasks. The simulation results show the high performance of the classification, and exhibits 97% correct classification ratio.