ISAR Imaging of Non-Uniform Rotating Targets Based on Optimized Matching Fourier Transform

Uniformly accelerated rotation of maneuvering targets can lead to quadratic phase terms, after translational motion compensation, which causes the cross-range signal in a certain range cell to be multicomponent linear frequency modulated (LFM) signals. Due to the time-varying Doppler frequencies of scatterers, ISAR image obtained by the conventional range-Doppler (RD) algorithm is blurred severely. To solve this problem, a new imaging method based on optimized matching Fourier transform (MFT) is proposed in this paper. In the new method, the equivalent rotation of targets is described by the rotation ratio (RR) which is the same for all scatterers on the target. The quadratic phase components caused by target's angular acceleration can be compensated by MFT, as soon as the RR is obtained. The estimation of RR is achieved according to the optimal imaging result with the image entropy as the cost function. As a result, the blurring of the ISAR image is eliminated effectively and a well-focused image is got. From the experimental results using simulated data, it can be concluded that the proposed method based on optimized MFT is effective in ISAR imaging of non-uniform rotating targets in terms of image quality and computational efficiency.