Steering vector random error modification robust beamforming

The steering vectors estimated by the existing beamforming methods generally depart from their ground-truth values when they are influenced by both steering deviation and random error. To solve this problem, a steering vector random error modification robust beamforming method is proposed in this paper. The steering vector is divided into both inside and outside parts of the pre-defined sector feature subspace, and then the robust beamforming for steering deviation is employed to estimate the component of the steering vector inside the pre-defined sector feature subspace. After that, the uncertainty set constraint is further added to estimate the component of the steering vector outside the pre-defined observation subspace, which then brings high accuracy in steering vector estimation. Simulation results indicate that the proposed method possesses a more stable signal-to-interference-plus-noise ratio and higher steering vector estimation accuracy than the existing robust beamforming methods under both steering deviation and random error. In addition, the peak of the steering vector beam response is the unbiased estimation of the target direction. Furthermore, sea-trial results illustrate the outstanding robustness of the proposed method under low signal-to-noise-ratio conditions.