Research on the position correction of component-level parametric scattering center models established in a forward approach

摘要

Summary form only given. This paper presents a position correcting method concerning the existing problem in the forward approach to establish component-level parametric scattering center models for known complex radar targets. The forward approach decomposes the target-model of radar into some parts and then extracts the scattering centers and their parameters through radar simulation. In the actual utilization of the forward approach, we found out that when there are more than one scattering source in a small region and they are close to each other, there are some differences between the SAR image reconstructed by the component-level parametric scattering center models established in a forward approach and the actual SAR image. Concerning the problem, this paper analyzes the reason. Comparing decomposing the target-model of radar into small pieces, the forward approach decomposing the target-model of radar into relatively large parts reduces data quantity but induces position error. So it is necessary to correct the positions. The whole procedure of the proposed method is composed of three parts: position locating, scattering sources distinguishing and position correcting. The method first locates the positions where the reconstructed SAR image and the actual SAR image have relatively large difference. Then, the scattering sources nearby the positions are distinguished by using the projection formulas. Finally, the positions of these scattering sources are modified slightly by running the phase correcting program to make the phase relationships between these scattering sources more reasonable. Thus making the reconstructed SAR image more similar to the actual SAR image. Results from position correction of some component-level parametric scattering center models show that the SAR image reconstructed by the corrected position parameter is closer to the actual SAR image than the uncorrected reconstructed SAR image in those corrected positions. This proves the validity of the proposed method.