Joint Method of ISAR Imaging and Scaling for Maneuvering Targets via Compressive Sensing

Generally, the inverse synthetic aperture radar images can be generated under the supposition that the targets rotate uniformly during a short observation. Some common methods functioning in this paper are then limited on image resolution and even have utter failure when the targets have more complex motions. In practical applications, it is difficult but significant to develop effective techniques of imaging and identification for maneuvering targets. To this end, this paper proposes a joint scheme to reconstruct the high-resolution images and estimate the rotational parameters for non-uniformly rotating targets. The electromagnetic images can be retrieved in the matched Fourier transform domain rather than Fourier plane, which also present great sparsity. Under the frame of compressive sensing, the rotational parameters and high-resolution image are solved alternately, during which the basis is determined by the parameters. This approach not only improves the image resolution but also reduces the noise. In addition, the enhanced image can further be cross-range scaled with the estimated rotational rate. The numerical results confirm the performance of the proposed method.