Abstract: A novel multi-sensor information fusion identification method is proposed for the low accurate rate of the transmitter individual identification caused by the various jamming signals and sensor error, which can enhance the stability and accurate recognition rate of the transmitter individual identification in the complicated environment. The proposed method integrates the Dempster-Shafer evidence theory and feature extraction to get the utmost out of feature information and decrease the influence of uncertain factors in the signal processing. The features are extracted from the detected signals. The self-adaptive fusion rule based on the decision vector is utilized to fuse the evidences transformed by features. The recognition results can be obtained by judgment rules. The simulation analyses of self-adaptive fusion rule and fusion identification method are performed, respectively. The results show that the self-adaptive fusion rule can achieve a great balance between computational efficiency and accurate identification rate. Compared with other identification methods, the proposed fusion identification method can provide more accurate and stable recognition results.

Key words: ordnance science and technology, emitter identification, information fusion, Dempster-Shafer evidence theory, feature extraction, self-adaptive evidence fusion