Investigation Of Neural Network Techniques For Anti-Stealth Radar Networks

This paper studies the data fusion technique of anti--stea1th radar networkand applies the neura1 network technique to the data fusion. The paper buildsup one simp1y RCS mode1 of the stea1th aerocraft, and study the system ofanti--stea1th radar network and the col1ocation form of the radar. Thefo11owing works are focused on this paper.l. Based on the viewpoint of combination optimization j this paperana1yses the similitude of data association of the single radarmu1titarget and the TSP, presents a new Hopfield energy function,and gives out one method based on the Hopfield network new dataassociation a1gorithm, then proves the new a1gorithm astringency.2. The paper presents one neural network model ---the seLf-organisedfeature-mapping mode1 (S0FM), and the data fusion algorithm of theminimum mean-square, then give out one seLf-organised data fusionalgorithm of the minimum mean-square.The conc1usion of the simu1ation indicates the upstanding performancesof these algorithm.