Automatic modulation classification based on machine learning

Automatic Modulation Classification (AMC) is a rapidly evolving technology, which can be employed in software defined radio structures, such as for military communication. Machine Learning can provide novel and efficient technology for modulation classification, especially for systems working in low Signal to Noise Ratio (SNR). For this work, a dynamic modulation classification system without phase lock is trialed. The signals are captured with different SNR and duration. Traditional Machine Learning based on the mathematical features is compared with Deep Learning based on the constellations. Based on these two methods, a hybrid model is provided. This model involved the novel Deep Learning at first and the feature classification as a supplement, which achieves good performance at low SNR area.