Classification of digital signals in cognitive radio based on second-order statistical approach

Abstract

An approach for multiclass digital signal classification based on second-order statistical features and multiclass Support Vector Machine (SVM) classifier is proposed. The proposed system is designed to recognize three different modulation schemes such as DPSK, PSK and MSK. The 2nd order cumulants of the real and imaginary parts of the complex envelope are extracted and these statistical features are given to multiclass SVM classifier for classification. The modulated signals are passed through the Rayleigh channel and Additive White Gaussian Noise (AWGN) channel before feature extraction. The evaluation of the system is carried on using 400 generated signals. The overall classification rate of the proposed system for various SNR levels is over 83%.