Modulation classification based on nonlinear functions and distances

In this paper, we propose a novel modulation classification algorithm based on high-order cumulants, and calculation of Euclidian distances. Non-linear transformation functions are also introduced to change the characteristics of the signals for calculating the multi-dimensional features. Simulation results are presented to demonstrate the superior performance of the proposed scheme compared with the existing hierarchical scheme. The averaged improvement for three different sample sizes is at least 18% over an SNR range of -5dB to 10dB of SNR for the four-class problem.