Individual Communication Transmitter Identification Using Support Vector Machines with Kernels for Polyspectrum of Box-Dimension

Abstract

The box dimension, as one of the fractal dimensions, can represent the discrepancies of different signals emitted by individual communication transmitters. However, the results of the box dimension's identification are usually dissatisfied. In order to enhance the clustering of this character, this paper presents a method using support vector machine with polyspectra kernels of the box dimension for classification of individual transmitters. The experiments indicate with the algorithm mentioned above. It shows that this method can improve the recognition rate.