Abstract:

A new automatic digital modulation recognition algorithm based on combined feature parameters is proposed for the application under the low signal-to-noise ratio (SNR). The feature parameters picked up from high order cumulants, instantaneous information and constellation characters of the signals are used as the classification vectors. The method can identify four classes of digital signals which are M-ary amplitude shift keying (MASK), M-ary frequency shift keying (MFSK), M-ary phase shift keying (MPSK) and M-ary quadrature amplitude modulation (MQAM) by using a back propagation (BP) neural network with the resilient back-propagation (RPROP) training algorithm as the classifier. The computer simulations show that this proposed algorithm effectively improves the practicability because of an overall success rate of 97% at the SNR of -2 dB.