Modeling And Simulation Of Airborne ISAR Imaging Of Aerial Targets

The main idea behind the concept of ISAR(Inverse Synthetic Aperture Radar) imaging is to exploit Doppler information induced by relative movement between the target and radar. And ISAR imaging provides an important way for non-cooperative target recognition, which has attracted broad attention; however the case where both the target and the radar are moving has not been intensively studied. Combined with the growing importance of air combat, this case shows broad prospects.Based on airborne radar, this dissertation investigated ISAR imaging of aerial targets. For relative motion model under this special case, a global coordinate is established. And following the general ISAR imaging process, resulting difficulties distributed in every step were thoroughly analyzed : translational and rotational compensation algorithms were generalized under such circumstance, furthermore autofocus algorithms were in-depth studied. For the dilemma of determining a stable imaging plane, which results from the target’s own kinematics, an optimization method was applied to solve the consequential image blurring. In this dissertation, concrete work and innovation points are summarized as follows:1. First, three coordinate systems put forward to model movements for this special case. Update coordinate values in the global coordinate system by the variation of the motion parameters, then multiply rotation matrix with coordinate values in their own coordinate to interpret induction motion, the data that has the characteristic of this special case can be generated. Also the reason for the difficulty in choosing the imaging plane in this case, due to time-varying effective rotation vector induced by the target rotation,was explained in chapter one.2. For this special case, motion parameter estimation model was set up and five translation compensation algorithms were compared and improved. Among which, distance and Doppler centroid method was investigated thoroughly. Owing to irregular phase component caused by complex relative movement, after phase compensation using centroid method, another refining process needed to improve the translational compensation, which took up a bit of the subsequent rotation compensation burden;3. For the difficulty induced by large rotational angle of this special case, an improved polar reformatting algorithm(PFA) was applied to increase image focus, which first autofocus the data after translational compensation, and then select the optimal interpolation method for classical PFA. The evaluation of the improved algorithm was verified by simulation according to different movement variables.4. When the similarity of respective motions(the target and the radar) is not high, relative motion is extremely complex. For this particular case, the imaging plane is unstable. To circumvent this problem, an algorithm based on image quality index and autofocus algorithm was presented in chapter 6 to select the time window, determining the best imaging plane; this solution reduced the above dilemma to an optimize issue. A double linear search is applied to determine the window time length and the central time instant, the imaging results of simulations demonstrate the effectiveness of the proposed algorithm.5. Matlab GUI was designed to implement ISAR imaging simulation system on the special case. The system can set up radar parameters and target scenes online, and can choose different imaging algorithms for variable aerial targets.