Modulation classification of MQAM signals based on density spectrum of the constellations

Blind modulation classification (MC) is an intermediate step between signal detection and demodulation, and plays a key role in various civilian and military applications. In this paper, a novel algorithm based on the density spectrum of digital signal constellations is presented. First we propose a new carrier estimation method (iterative approximation algorithm of carrier estimation), which can acquire the accurate approximation value of the carrier frequency. Then we perform the wavelet denoising based on wavelet transform modulus maximum. Finally, the discriminating feature is derived from the density spectra to identify the different modulated signals. The density spectra based classifier is much simpler and has lower computational complexity than the traditional classifier based on constellation reconstruction. The simulation results show the relatively high efficiency and recognition accuracy of the method.