Robust adaptive beamforming for coprime array based on subspace orthogonality

This paper presents a robust adaptive beamforming (RAB) algorithm for coprime array based on subspace orthogonality. Initially, we eliminate residual noise to improve the accuracy of the noise power estimation. The Desired Signal Steering Vector (DSSV) is derived from the noise-free desired Sample Covariance Matrix (SCM). To obtain a virtual Uniform Linear Array (ULA) much longer than the actual linear array, we derive the continuous elements in the difference coarray and construct the Toeplitz matrix to create the virtual SCM. The interference Steering Vectors (SVs) are optimized by exploiting the orthogonality between the complementary subspace and the actual interference SVs. Subsequently, we accurately reconstruct the Interference-plus-Noise Covariance Matrix (INCM) in the physical domain, utilizing the maximum correlation principle, and derive the optimal weight vector for beamforming. Through simulation results, we demonstrate the robustness of the proposed algorithm in dealing with look direction errors, incoherent local scattering errors, array element gain and phase perturbation errors.