Supervised modulation classification based on ambiguity function image and invariant moments

Automatic modulation classification (AMC) has been a significant research topic in communication systems especially cognitive radio systems. The development of AMC algorithms is still at an immature stage for practical applications. In this paper, a supervised modulation classification scheme is proposed for automatic recognition of different types of communication signals. The supervised classification scheme is based on the distinction of ambiguity function (AF) images of different modulation signals. Two sets of classification feature vectors are exploited from the AF image. One feature vector is a low-dimensional vector by using the principal component analysis (PCA) technique on the AF image. The other feature vector is obtained by computing the invariant moments (IMs) of the AF image due to the different shape information of AF images. Based on the extracted features, the final classification is accomplished through the support vector machine (SVM) classifier. The proposed algorithm is capable to recognize seven different modulation signals: ASK, PSK, QAM, FSK, MSK, LFM and OFDM. Final experimental results demonstrate the efficiency and the robustness of the proposed algorithm in low SNR situations.