ISAR Imaging for Micromotion Targets Using the Hierarchical Bayesian Prior

Lin Sun,Tianyun Wang,Xinfei Lu,Weidong Chen
DOI: https://doi.org/10.1109/apsar.2015.7306268
2015-01-01
Abstract:The micromotion of structures in a target may introduce a micro-Doppler (m-D) effect in the inverse synthetic aperture radar (ISAR) imaging, which contaminates the main body image of the classical range Doppler algorithm (RDA). This paper proposes a novel imaging method for micromotion targets, which assigns a hierarchical Bayesian prior to the element in the range profile, for the purpose of promoting the sparsity and maintaining the continuity pattern along the slow time dimension. To overcome the intractability of the posterior distribution, the Gibbs sampler is utilized to achieve the Bayesian inference. Subsequently, the range profile can be estimated from the expectation of the conditional posterior distribution, then a clear ISAR image can be obtained after cross compression. Experimental results validate the performance of the proposed method.
What problem does this paper attempt to address?