Super-resolution ISAR Imaging Method with Sparse Statistics

Zheng Bao
2012-01-01
Journal of Xidian University
Abstract:To improve the resolution of inverse synthetic aperture radar(ISAR) imagery with short observation,this paper proposes a super-resolution method for ISAR imaging with sparse statistics.Combined with strong sparsity of ISAR images,the imaging process is approximately modeled by statistical probability distributions.By maximum posterior probability and the Bayesian estimation approaches,the close form of sparsity control parameters is deduced,and the optimal image is solved by the conjugate gradient method.Also,with the CFAR(constant false alarm rate) and bandwidth extrapolation technique involved in a stage-by-stage procedure,the robustness of parameter estimation and imaging is improved.Real-data experiments validate the superiority of the proposed method.
What problem does this paper attempt to address?