Abstract—Air targets threat assessment is an important part of command and control. Recently, the technical difficulties of assessment are how to deal with quantitative and qualitative information effectively, and how to analyze and process uncertain information. Owing to the singleness and the insufficiency of traditional processing methods, a new parallel processing method is put forward. In order to get the dynamic threat value, the strong generalization ability of Back-Propagation is used to deal with quantitative indicators, at the same time, the strong reason ability of Bayesian network is used to analyze qualitative indicators to acquire the static threat value. Finally, we get comprehensive threat value by linear weighting the two kinds of threat values. Through the case analysis and the prediction of ten batches of target, combined with the simulation in the Matlab and Genie software, the method is proved to be effective. The result of the simulation shows: the comprehensive threat value is more credible than that got by traditional method.


Index Terms—Threat assessment, Back-Propagation netwrok, Bayesian network.