A What-and-Where fusion neural network for recognition and tracking of multiple radar emitters

A neural network recognition and tracking system is proposed for classifica tion of radar pulses in autonomous Electronic Support Measure systems. Rada r type information is considered with position-specific information from ac tive emitters in a scene. Type-specific parameters of the input pulse strea m are fed to a neural network classifier trained on samples of data collect ed in the field. Meanwhile, a clustering algorithm is used to separate puls es from different emitters according to position-specific parameters of the input pulse stream. Classifier responses corresponding to different emitte rs are separated into tracks, or trajectories, one per active emitter, allo wing for more accurate identification of radar types based on multiple view s of emitter data along each emitter trajectory. Such a What-and-Where fusi on strategy is motivated by a similar subdivision of labor in the brain. The fuzzy ARTMAP neural network is used to classify streams of pulses accor ding to radar type using their functional parameters. Simulation results ob tained with a radar pulse data set indicate that fuzzy ARTMAP compares favo rably to several other approaches when performance is measured in terms of accuracy and computational complexity. Incorporation into fuzzy ARTMAP of n egative match tracking (from ARTMAP-IC) facilitated convergence during trai ning with this data set. Other modifications improved classification of dat a that include missing input pattern components and missing training classe s. Fuzzy ARTMAP was combined with a bank of Kalman filters to group pulses transmitted from different emitters based on their position-specific parame ters, and with a module to accumulate evidence from fuzzy ARTMAP responses corresponding to the track defined for each emitter. Simulation results dem onstrate that the system provides a high level of performance on complex, i ncomplete and overlapping radar data. (C) 2001 Elsevier Science Ltd. All ri ghts reserved.