This study proposes a robust wideband beamforming algorithm based on spectrum separation at divided subspaces to address the possible sensors array manifold deviations in practice. In this algorithm, an efficient sensors array manifold calibration approach based on subspace division is put forward and an improved interference-noise covariance matrix reconstruction approach based on spectrum separation is derived. The algorithm first divides the spatial space into several subspaces by Capon

more »

... al spectrum and then calibrates array manifold deviations in these subspaces. Subsequently, the noise and interference information is extracted accurately after spectrum separation. Based on the extracted information, the optimal beamformer can be finally designed. Simulation results prove the excellent performance of the proposed algorithm in the presence of various array manifold deviation types and extents.