An Adaptive Sparse Constraint ISAR High Resolution Imaging Algorithm Based on Mixed Norm

Abstract

Based on the sparsity of inverse synthetic aperture radar (ISAR) signal, in this paper, a novel high resolution imaging algorithm is proposed. In this method, an optimal ISAR signal model based on mixed norm is established by using compressed sensing theory. The high-resolution ISAR image with short coherent accumulation time is realized by solving the optimization model. The main advantages of the proposed approach are: The model makes use of the l2,0 mixed norm to realize faster convergence and improve the computational speed of the model solution obviously. Moreover, according to the result sparsity of each iteration under arbitrary noise, the regularization coefficient in the model can be adjusted adaptively, which avoids the complex process of repeated attempts, otherwise, the optimal coefficient needs to be estimated and attempted by the statistical characteristics of the noise and signal. The effectiveness of the proposed method is verified by simulated and measured data.