Moving Target Tracking Based on Improved GMPHD Filter in Circular SAR System

Abstract

A new moving target tracking method based on an improved Gaussian mixture probability hypothesis density (GMPHD) filter in a circular synthetic aperture radar (SAR) system is proposed. In order to obtain the temporal dynamic behavior of moving targets in a SAR scene, a multiple-target observation is first extracted from a sequence of clutter-suppressed subaperture circular SAR images. Then, a new multiple-target tracking model for the circular SAR system is established based on the dynamic behavior of moving targets. An improved labeled GMPHD filter with an adaptive target birth intensity is proposed to realize the moving target detection and tracking in a strong ground clutter environment, which is suitable for the circular SAR system. This method can not only provide accurate estimates for the number and states of moving targets but also can extract their trajectories. The effectiveness of the proposed method is verified through simulations and experimental Gotcha data.