Automatic modulation classification using support vector machines and error correcting output codes

A new approach, which because of support vector machines (SVMs) and error correcting output codes (ECOC), is proposed for automatic digital modulation classification (MC). Combining statistics and spectral feature sets, SVM classification and ECOC techniques are used to improve the generalization ability of classifiers. In contrast to the popular artificial neural networks (ANN) approach, this method avoids the problem of local optimum and requires less training data. Simulation results show that the proposed method is more efficient than other methods in low signal-to-noise ratio.