Open AccessAutofocus technique for radar coincidence imaging with model error via iterative maximum a posteriori
Author(s) -
Zhang Feng,
Liu Xunling,
Zhou Xiaoli,
Wang Xu,
Liu Weijian
Publication year - 2019
Publication title -
the journal of engineering
Language(s) - English
Resource type - Journals
ISSN - 2051-3305
DOI - 10.1049/joe.2019.0129
Subject(s) - autofocus , computer science , coincidence , maximum a posteriori estimation , deconvolution , algorithm , radar , radar imaging , artificial intelligence , iterative method , computer vision , mathematics , maximum likelihood , optics , physics , statistics , medicine , telecommunications , alternative medicine , pathology , focus (optics)
Radar coincidence imaging (RCI) is a recently developed staring imaging technique based on the wavefront random modulation and temporal–spatial stochastic radiation field. Before coincidence imaging processing, the RCI needs to compute the reference matrix accurately. Unfortunately, model error usually exists, which degrades the imaging performance considerably. In this article, by exploiting the sparse prior of target, the authors propose a sparsity‐driven autofocus imaging via iterative maximum a posteriori (SA‐MAP) algorithm for the RCI when model error exists. The algorithm performs well via simultaneous sparse imaging, self‐calibration and parameters update. Simulation results demonstrate the validity of the proposed algorithm and performance improvement over the existing algorithms.