Digital Signals Classification in Cognitive Radio Based on Discrete Wavelet Transform

摘要

In this paper, Discrete Wavelet Transform (DWT) based digital signals classification is proposed. First, the modulated signals are decomposed by using DWT. Secondly, a set of biggest wavelet coefficients is selected for training the classifier. Thirdly, a supervised classifier system based on SVM is constructed to classify the modulation scheme of the unknown signal. The modulation schemes used in the proposed systems are DPSK, PSK and MSK. The modulated signals are passed through an Additive White Gaussian Noise (AWGN) channel before feature extraction. 400 generated signals are used to evaluate the proposed system. The maximum classification rate achieved by the proposed system is 75% to 97% while using 30% biggest wavelet coefficients.