Active blanket jamming identification method based on rough set and decision tree

Abstract

Aiming at the classification or identification problem of active blanket jamming, a jamming identification method is proposed, which combines the upper approximation set, lower approximation set, border set theory of rough set and decision tree. The method first extracts the time-domain features of the signals, and divides the training set according to the degree of noise impact of the sample to train the decision tree respectively; then, the noise detection of the jamming signals is carried out by using the rough set theory; finally, the identification of three typical jamming patterns is realized through the decision tree. The comparison of simulation experiments shows that compared with the traditional method, this method has higher recognition accuracy, good real-time performance and noise robustness.