Research On ISAR Motion Compensation Methods Based On Time-Frequency Analysis

Inverse synthetic aperture radar (ISAR) imaging is an effective way to acquire high resolution images of targets of interest at long range and as such is an irreplaceable tool in the task of non-cooperative target recognition of both ships and aircraft.One significant problem with ISAR image formation is the assumption of time invariance of the Doppler frequency used to resolve the image in the cross-range. Timevariant Doppler becomes present in an ISAR signal when an aircraft is maneuvering or a ship is pitching and rolling during the coherent processing interval and is typically referred to as motion error. Because of this motion error and the fast Fourier transform's (FFT) basic assumption of time-invariance Doppler, the use of the FFT to resolve the image in the cross-range will cause extensive blurring and leave the image unrecognizable.One proven method of motion compensation is the adaptive joint time-frequency (AJTF) algorithm. However, this algorithm is not without two significant weaknesses of its own. The first problem is that the computational burden of the exhaustive search used to extract the motion compensation parameters is quite large which limits its usefulness in an operational situation. The second problem with the AJTF algorithm is that the target must follow the mathematical model of the AJTF which, in particular, states that rotational motion must be confined to a 2D plane (i.e., no 3D motion).Using the simulated ISAR data of a triangle wing target, two new AJTF algorithms were researched in this thesis, which respectively use Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) to take the place of exhaustive search for the extraction of the motion compensation parameters,and lagely speed up the convergence of the algorithm. And prove that, the PSO algorithm,because of its inherent mechanism,is more effective for ISAR application.Based on the AJTF-PSO algorithm designed above,we introduced a 3D motion detection algorithm,which was able to sort good imaging intervals in the real radar data set from poor imaging intervals which violated the 2D motion assumption of the mathematic model. It was shown that the imaging interval indicated as possessing a low degree of 3D motion created well-focused images using the present motion compensation algorithms. Imaging intervals that were indicated as possessing a high degree of 3D motion did not focus even after being motion compensated. These imaging intervals can now be discarded instead of presented to the ISAR operator.The first chapter of the thesis introduced the phylogeny of ISAR and the present researches both at home and abroad .Chapter 2 introduced the principle of ISAR imaging, and compare the main conventional motion compensation algorithms by simulated data.Chapter 3 introduced the AJTF motion compensation algorithm using evolutionary techniques. Chapter 4 introduced a 3D target motion detection technique and the automatical selection of radar data ;In chapter 5, conclusions and future work are explained.