A Method Based on Equivalent Measurement of Radiation Fields for Coded-Aperture Imaging With System Errors

Coded-aperture imaging (CAI) is a promising radar imaging technology, which has the advantages of high resolution, forward-looking, and staring imaging. However, it still faces some challenges. In the actual imaging process, there are many kinds of errors, such as modulation error of the coded antenna, off-grid error, and imaging distance error, and these errors will affect the accuracy of the radiation field and the derivation of the imaging model. When the actual imaging model is inconsistent with the derived imaging model, the imaging result may deteriorate seriously or even fail to be imaged. In addition, the imaging efficiency needs to be improved urgently. Therefore, we propose a data-driven method based on equivalent measurement of radiation fields to tackle these problems. First, a new radiation field modeling method is proposed, which takes into account the accuracy and complexity of radiation field modeling. It can greatly improve the reconstruction quality of sparse and extended targets under system errors, and it also has the advantage of low workload and easy implementation. Then, an artificial neural network algorithm based on equivalent measurement of radiation field (ANN-EMRF) is designed to improve imaging efficiency. Simulation and laboratory experiments are carried out to demonstrate the feasibility of the proposed method under various system errors.