Modulation classification in cognitive radios for satellite and terrestrial systems

In order to realize improving signal detection probability using interference canceler and multi-system detection techniques in spectrum sharing environment in cognitive radio of satellite and terrestrial system, it is required to classify the digital modulation type in an environment without handshaking between the transmitter and receiver. The previous methods can hardly classify modulation type precisely unless high SNR (signal power to noise power ratio) is provided, which is barely possible to meet in satcomm downlink. To solve this problem, this paper proposed a novel method to achieve the classification of QAM and PSK modulation, which are popular modulation types being used for satellite communications. The features of phase and amplitude are extracted by Pseudo Wigner-Ville time-frequency distribution. With the feature obtained above, simulation results indicated that this method offers an accurate classification.