Modulation classification in alpha-stable noise

In this paper, we propose a modulation classifier for non-Gaussian noise environment. The modulation model is formed via autoregressive spectrum modeling, where the additive noise is assumed to obey symmetric alpha-stable distribution. The modulation model used the instantaneous frequency and bandwidth parameters as obtained from the roots of the autoregressive polynomial. The distinguish features are extracted based on fractional lower order moments of alpha-stable distribution random variable. Computer simulations demonstrate its efficiency and comparison with other methods is also given.