Automatic Modulation Classification using combination of Wavelet Transform and GARCH model

Automatic Modulation Classification (AMC) is an important step before demodulation. This process significantly helps to the receiver in recognition which has no, or limited, information of received signals. Nowadays, AMC plays an important role in many applications such as spectrum management, cognitive radio, intelligent modems, surveillance, and interference identification. This paper evaluates the effectiveness of the combination of the Generalized Autoregressive Conditional Heteroscedasticity (GARCH) model with the Discrete Wavelet Transform (DWT) for AMC. In the proposed method, at first, WT is applied on the received data samples. Our exact analysis indicates that the wavelet coefficients have heteroscedasticity property and GARCH model is appropriate to represent them. The parameters of GARCH model are extracted as the features and are applied to the support vector machine (SVM) classifier to determine the modulation type and constellation size simultaneously. We consider six different types of digital modulation schemes including, phase shift keying (PSK) and quadrature amplitude modulated (QAM). The performance of the proposed method in non-fading and fading channels in the presence of Gaussian noise is evaluated. The results indicate the superior performance of the proposed method in comparison with the recently introduced methods.