Removing Micro-Doppler Effect in ISAR Imaging by Promoting Joint Sparsity in Range Profile Sequences and the Time-Frequency Domain

Abstract

The micro-Doppler effect induced by rotating parts interferes the inverse synthetic aperture radar (ISAR) imaging of the target’s rigid body heavily. To extract useful signals from rigid body and remove the micro-Doppler interference, this paper proposes a novel approach by promoting joint sparsity of the rigid body both in range profile sequences and time-frequency domain. The problem is formulated and modeled as a convex optimization with double joint sparsity penalty, and the exact solution can be obtained after multiple alternate iterations. In addition, the paper analyzes the effect of the scatterer’s linear migration through resolution cells (MTRC) on micro-Doppler removal. In order to avoid the signals from migrated rigid scatterers being removed together with the micro-Doppler signals, this paper proposes another improved reweighted method, in which the weights are designed to correlate slices with same frequency in neighboring range cells, to preserve migrated rigid scatterers and remove micro-Doppler interference meanwhile. Experimental results based on both synthetic and Electromagnetic simulated data demonstrate the validity of the proposed methods.