Radar Cross Section Identification of Air Targets using the Cosine Transform and Neural Networks

Abstract

Target identification has always been as important as the radar itself. This paper presents an innovative approach to the problem of aircraft identification from its radar cross section. Radar cross section (RCS) data are passed through Discrete Cosine Transform. The Cosine Transform extracts the dominant features of the data. This provides compression of large amounts of RCS information into much smaller sizes of significant information. Therefore the Neural Network Classification Algorithm runs much faster with respect to the case where the original data are used as the input. In this study, patents relevant to the topic have also been discussed. Keywords: Radar Cross Section Identification, Cosine Transform, Neural Networks, radar cross section, Fourier transformation techniques, secondary radiation, scattering, BACKSCATTERING SIMULation, Artificial Neural Network, DCT matrix, neural classifier, regression, Mean-Squared-Error, propagation, windowing technique, Discrete cosine transform, identification