Modulation classification by means of different orders statistical moments

This paper deals with modulation classification of 4PSK (4 states phase shift keying) and 16QAM (16 states quadrature amplitude modulation) using a pattern recognition approach. The discriminating feature is build as an optimized combination of fourth and second order moments in order to maximize the probability of correct classification (P/sub /spl infin//). This system is an extension of the one proposed in Boitreau and Le Martret (1996) where the discriminating feature is fourth order cumulant slice. Theoretical and experimental performances are presented for different signal to noise ratios (SNR) and are compared to the one obtained in Boitreau and Le Martret. A P/sub /spl infin//=0.99 is reached for SNR=0 dB and only 500 transmitted symbols.