Blind digital modulation classification based on M-TH power nonlinear transformation

Automatic Modulation Classification (AMC) has received a major attention last decades, as a required step between signal detection and demodulation. In the fully-blind scenario, this task turns out to be quite challenging, especially when the computational complexity and the robustness to uncertainty matter. AMC commonly relies on a preprocessor whose function is to estimate unknown parameters, filter the received signal and sample it in a suitable way. Any preprocessing error inherently leads to a performance loss. To improve the robustness of the blind AMC, we propose to proceed almost directly on the received signal - with neither matched-filtering step nor synchronization step. In this paper, Analytical M^th-Power nonlinear Transformation (AMPT) is considered for its robustness towards timing, phase and frequency uncertainty. The generated feature-vector then feeds a Minimum Distance classifier. Numerical simulations show the effectiveness of the proposed method for a 7-class problem of low-order modulations.