Combining Genetic Algorithm Based Joint Time-Frequency Analysis and Keystone Transform for ISAR Image Enhancement

Abstract:
Translational and rotational motion of maneuvering targets can cause range and cross-range cell migrations, culminating in a highly smeared final inverse synthetic aperture radar (ISAR) picture. To overcome this problem, a method for ISAR high-resolution imaging and motion compensation (MOCOMP) is proposed that combines adaptive joint time-frequency (AJTF) analysis and keystone transform (KT) based on genetic algorithm (GA). In synthetic data processing, the motion characteristics of the target are estimated using the Matching Pursuit (MP) technique with a third-order model of polynomial function. The Gabor-wavelet transform (GWT) is utilized as a JTF tool to generate a corresponding 3D time-range-Doppler cube. Keystone transform as remapping of the time axis is applied and Translational Motion Compensation (TMC), Rotational Motion Compensation (RMC), and KT have all been proved to operate well together. In raw data processing, the high-resolution image is obtained by using an AJTF and KT based on a genetic algorithm sans MP and GWT to compensate for migrations in the range and Doppler axis.