Ship formation recognition based on information fusion of spaceborne IMINT and ELINT

Abstract

A method of ship formation recognition is proposed based on the information fusion of spaceborne IMINT (image intelligence) and ELINT (electronic intelligence) in this paper. Firstly, the composition and the battle array of the observed ship formation are derived from the spaceborne IMINT information of each individual ship. The beliefs about the composition and the battle array of the observed formation are calculated separately. A synthetic evaluation method is used to get the BPAF (basic probability assignment function) from IMINT information using the aforementioned two beliefs. Secondly, by computing the matching measure between the observed emitter set and that of the typical formation entity, the BPAF based on ELINT information is obtained. Thirdly, the two BPAFs based on the information of spaceborne IMINT and ELINT are combined with DS (Dempster-Shafer) evidence theory. Finally, a decision is made according to the combined BPAF. The experiment indicates that our proposed methods to obtain the BPAFs are practicable and the formation recognition accuracy is greatly improved compared to the results which use only one of the two sources.