Individual Communication Transmitter Identification Using Correntropy-based Collaborative Representation

In this paper, an efficient radio transmitter identification method is proposed for identifying radio transmitters. The square integral bispectra (SIB) transformation is firstly utilized to extract the features from the raw signal data of radio transmitters, which recasts the problem of communication transmitter identification into the form of measuring similarity between points in its metric space. Then we use the collaborative representation framework as a platform to develop a novel classification model, correntropy-based collaborative representation classifier (CECRC), for identifying different radio transmitters according to the similarity between the points in the SIB feature space. Extensive experimental results on real-world data sets demonstrate the effectiveness of our proposed method.