Superresolution Imaging for Forward-Looking Scanning Radar with Generalized Gaussian Constraint

A maximum a posteriori (MAP) approach, based on the Bayesian criterion, is proposed to overcome the low cross-range resolution problem in forward-looking imaging. We adapt scanning radar system to record received data and exploit deconvolution method to enhance the real-aperture resolution because the received echo is the convolution of target scattering coefficient and antenna pattern. The Generalized Gaussian distribution is considered as the prior information of target scattering coefficient in MAP approach for the reason that it could express different target scattering coefficient properties with the control of statistic parameter. This constraint term makes the proposed algorithm useful in different applications. On the other hand, the reconstruction problem can also be viewed as the lp-norm (0 < p ≤ 2) regularization. Simulation results show the robustness of the proposed algorithm against additive noise compared with other superresolution methods.