Clutter estimation based on compressed sensing in bistatic MIMO radar

In this paper, the clutter estimation problem in the bistatic multiple-input and multiple-output (MIMO) radar system is considered, and a novel compressed sensing (CS)-based model is proposed to describe the clutter by exploiting the clutter sparsity in the angle domain. Then, the CS-based methods are adopted to reconstruct the sparse clutter, and to estimate the clutter scattering coefficients and angles. Additionally, different from the traditionally colocated MIMO radar system with the antenna distance being half of wavelength, we show that the optimal antenna distance in the CS-based radar system can be obtained by minimizing the mutual coherence of the dictionary matrix. Moreover, since the sparse reconstruction performance depends on the geographical positions of the clutter scatterers, an indirect method based on the mutual coherence is proposed to measure the estimation performance, and to optimize the radar parameters. Simulation results show that the CS-based method can estimate the clutter information efficiently, and the better estimation performance is achieved by optimizing the radar parameters.