Classification using a novel support vector machine fuzzy network for digital modulations in satellite communication
Dan Wu, Xuemai Gu, Qing Guo · 2005
To make the modulation classification system more suitable for signals in a wide range of signal to noise rate (SNR), a novel support vector machine fuzzy network (SVMFN) is presented in this paper. The SVMFN employs a new definition of fuzzy density which incorporates accuracy and uncertainty of the classifiers to improve recognition reliability. Further, three efficient features with high robustness and less computation are extracted from intercepted signals to classify eleven digital modulation types (i.e. 2ASK, 4ASK, 2FSK, 4FSK, 2PSK, 4PSK, 8PSK, 16QAM, TFM, /spl pi//4QPSK and OQPSK). Computer simulation shows that the proposed scheme has the advantages of high accuracy and reliability (success rates are over 97.5% when SNR is not lower than 0 dB), and adapt to engineering applications.