A New Method for MQAM Signals Identification

The paper is presented a new method to directly classify various QAM signals in low SNR in the intermediate frequency (IF). The method extracts the amplitude of signal after wavelet transform using the optimal scale to get the maximum gain and to select the clear features in the distribution centers in the histogram for identifying different MQAM signals, avoiding the calculation error of frequency offset and phase bias caused by moving the signal from IF to baseband. The simulation results show that the correct identification accuracy is not less than 95% for 16 QAM, 32 QAM and 64 QAM signals until SNR = 3 dB, which have proved the validity of the method.