Radio Transmitter Classification using a New Method of Stray Features Analysis Combined with PCA

This paper introduces an efficient technique to design a classifier for identifying radio transmitters with the same model. A new method, square integral bispectra(SIB), is used to extract the unique stray features of individual transmitted signal, and the principal component analysis(PCA) method is utilized further to extract a low-dimensional classification vector. Then,the modulation parameters of individual transmitter significant to classification are combined with the low-dimensional stray features to form a single characteristic feature vector that can effectively represent the target of concern over a broad range of sample data. A support vector machine(SVM) based on Gaussian kernel-function is implemented to realize the individual transmitter classification. The experiments on FSK radio transmitters demonstrate that the suggested technique is highly accurate and robust even in the presence of excessive noise and can solve the problem of identifying individual transmitters with the same model and manufacturing lot.