Constellation-wavelet transform automatic modulation identifier for M-ary QAM signals

Automatic modulation identification (AMI) is a significant technique to identify the modulation type of a received signal, and it is of great importance in information war and communication countermeasure. In this paper, we propose a constellation-wavelet transform automatic identifier (C-WT AMI) to identify the order of the received QAM signal. QAM is a widely used modulation type due to its high spectrum efficiency. From the characteristics of QAM constellation, we perform Haar wavelet transform of the received signal with the most appropriate scale, and then calculate the ratio of the maximum and minimum value q̌M of the Haar wavelet transform magnitude of received signal, and we can finally determine the signal order M according to the relationship between q̌M and M. To compute the most appropriate wavelet scale, estimating carrier frequency and sampling frequency is necessary. In this paper, we propose a method to estimating sampling frequency in practical engineering implement. Considering that distinct peaks may occur in the Haar wavelet transform magnitude due to symbol phase changes, and it could impact the identification correct rate, so an effective algorithm to eliminate the influence of the peaks is also proposed. Simulation results show that the proposed scheme is of good performance, and it can almost be inerrable when SNR>4dB. Besides, compared to the methods proposed in other papers, our scheme shows a better performance with higher correct rate.