Dynamic ISAR Imaging Method for Multiple Moving Vehicles Based on OMP-CADMM

Abstract

Inverse synthetic aperture radar (ISAR) technology has attracted considerable attention in smart traffic monitoring, owing to the superiority of high-resolution imaging and tracking for moving targets in all-weather. However, in practical application cases, it is hard to obtain a globally-focused image of multiple vehicles by traditional ISAR imaging techniques due to the nonuniform motions of vehicles. Hence, a dynamic ISAR imaging method for multiple moving vehicles based on the joint of orthogonal matching pursuit (OMP) and consensus alternating direction method of multipliers (CADMM) is proposed in this paper. First, the preprocessing for raw data is performed in the wavenumber domain to simplify the imaging task into linear phase error estimation and distributed image recovery. Based on the sparsity of moving vehicles, the OMP method is utilized to estimate the dynamic motion phase error of consecutive frames data for different moving vehicles with less computational overhead. Then, the dynamic ISAR imaging problems of multiple moving vehicles can be transformed into the distributed imaging problem with variously estimated phases. Motivated by this fact, the CADMM optimization framework is exploited for the distributed ISAR imaging problem to obtain a globally-focused result according to the consensus constraint characteristic of multiple vehicles. Lastly, both the simulation and real data experiments are conducted to verify the performance and feasibility of the OMP-CADMM method. Furthermore, the experimental results show that the proposed method is capable of dynamic imaging and de-noising for multiple vehicles without signal separation.