Radar emitter threat evaluation based on the algorithm involving behavioral characteristics and BiasSVD

Abstract In order to minimize the impact of errors and uncertainties that arise from signals received by reconnaissance equipment and airborne radar on emitter threat evaluation, to reduce such evaluation’s dependence on parameters, and to predict better the extent to which an air-battle target is posing a threat, the paper examined the radar emitter and dynamic characteristics of a target from a behavioral perspective using data-fusion-based emitter threat evaluation. It was followed by a shift from the concrete description of a complicated air-battle situation to an abstract one, and the proposed algorithm’s fault tolerance increased. With that, a threat assessment system built on behavioral characteristics was established. Under that system, we calculated the membership of each sub-behavior indicator’s vague dataset, optimized threat evaluation weights using the dynamic variable weight method, and computed the threat value with an improved radar chart. This was how an emitter threat was swiftly and accurately assessed. As to failures to precisely evaluate the threat of a target due to a lack of needed information for it being beyond the scope of an aircraft warning device or a radar receiver, we, in this paper, employed the collaborative filtering algorithm and predicted the threat degree in the case of non-existence of emitter behavior or existence of a target escaping radar detection by analyzing the situational environment of other fighter aircraft in the same airspace.