Radar Forward-Looking Imaging for Complex Targets Based on Sparse Representation With Dictionary Learning

Radar forward-looking imaging has always been a difficult problem in the radar detection. Modulating the wavefront of radar transmitted signals provides a feasible method for radar forward-looking imaging. The existing high-resolution radar imaging algorithms assume that the target scattering coefficients are sparsely distributed. However, for complex targets, the scattering coefficients no longer satisfy the sparse prior. To solve the problems of forward-looking imaging for complex targets, in this letter, a sparse representation imaging method with dictionary learning is proposed. First, the principle and imaging model of microwave modulation are introduced to achieve radar forward-looking imaging. Second, the dictionary learning method is developed to learn adaptive transformation, which exploits the edge features and structural information as well as provides a sparser presentation to further improve the quality of radar images. Third, the imaging performance for different types of complex targets under different signal-to-noise ratios (SNRs) is analyzed. The simulation results show that the proposed method can effectively reconstruct different types of complex targets.