TARGET DISCRIMINATION USING NEURAL NETWORKS WITH TIME-DOMAIN OR SPECTRUM MAGNITUDE RESPONSE

Several different memory-based neural networks are used to discriminat e radar targets based on their early-time, aspect-dependent response. The beginning of the response is difficult to locate in practice, so w e use only the magnitude of the time response's DFT Spectrum as input to the neural network, thus eliminating time-shift, uncertainty. Espec ially promising is the Recurrent Correlation Accumulation Adaptive Mem ory-Generalized Inverse (RCAAM-GI) cascade neural network. From the si mulation results, the network demonstrates a decision strategy which i s flexible, parallel adaptive, computation space efficient, and highly noise tolerant. Performances of the networks presented in this paper are compared with those of existing networks.