Individual Communication Transmitter Classification of Weighted Feature Set Based on RST

Feature extraction for individual communication transmitter identification is one of the major issues in the identifying process. A neighborhood rough set is proposed in this paper, in order to search for the good feature subset. Then we present a SVM classification approach of weighted feature set based on the significance of an attribute. The result of experiments shows that the reduced feature subset is acquired ,and the classification accuracy of weighted feature subset is much better than that of non-weighted feature subset.