MPSK signals modulation classification using sixth-order cumulants

New features based on sixth-order cumulants are proposed for classification of MPSK signals. Theoretical analysis justifies that the ability of suppressing multi-path fading improves as as the order number of cumulants features increases. Performance of proposed approach is demonstrated via computer simulations. The multi-path channel is assumed to be FIR model. MPSK signals are corrupted by Gaussian noise. The simulation results of probability of correct classification show that sixth-order cumulants method work better than fourth-order cumulants under the deeply multi-path fading channel environments and the fourth-order cumulants method work better than sixth-order cumulants when the signal-to-noise ratio (SNR) is lower than or equal to 0dB.