Feature extraction of radar targets using Evolutionary Adaptive Wavelet Transform

In this paper, the evolutionary adaptive wavelet transform (EAWT) is applied to the feature extraction of radar targets. The two features which are important for NCTR are scattering centers and resonance frequencies. EAWT algorithm use evolutionary programming (EP) for the time-frequency parameter extraction instead of FFT and the bisection search method used in conventional adaptive wavelet transform (AWT). Therefore, the EAWT has a better performance than the conventional AWT. In the simulation using wire target (Airbus-like) and simulated MIG-25, the comparisons with the conventional AWT are presented to show the superiority of the EAWT algorithm in the feature extraction of radar targets. The EAWT can be effectively applied to the radar target recognition.