Cyclostationary Features Based Modulation Classification in Presence of Non Gaussian Noise Using Sparse Signal Decomposition

Abstract

Automatic modulation classification�(AMC) is a salient component in the area of cognitive radio, signal detection, interference identification, electronic warfare, spectrum management and surveillance. The majority of the existing signals detection and classification methods presume that the received signal is corrupted by additive white Gaussian noise. The performance of the modulation classification algorithms degrades severely under the non-Gaussian impulsive noise. Hence, in this paper, we introduce a robust algorithm to identify the modulation type of digital signal�contaminated with non-Gaussian impulse noise and additive white Gaussian noise (AWGN) using a sparse signal decomposition on hybrid dictionary. The algorithm first detects and removes the impulse noise using sparse signal decomposition thereafter it classifies the modulation schemes using cyclostationary feature extraction algorithm. Simulation results demonstrate the superiority of the proposed method under different non-Gaussian impulse noise and AWGN conditions. The performance of the proposed classifier is evaluated using well known classifiers available in the literature. � 2017, Springer Science+Business Media New York.