Skip to Main Content
This paper presents a method of using support vector machine (SVM) with Second Generation Wavelet kernel (2G Wavelet kernel) for classification of individual transmitters working in frequency-hopping spread spectrum (FH-SS) modulation. Simulation results show that this method can significantly improve the efficiency of the identification in comparison with those based on First Generation Wavelet kernel (1G Wavelet kernel). Meanwhile, it can also achieve a satisfactory classification rate. Furthermore, the relationship between the classification rate and SNR is also discussed.