An Analytically Derived Robust Adaptive Beamformer via Interference-plus-noise Covariance Matrix Reconstruction
Jiajun Yan, Yue Ivan Wu, Huawei Chen · IEEE Transactions on Aerospace and Electronic Systems · 2025
The performance of the conventional sample matrix inversion (SMI) beamformer significantly degrades due to mismatches between the actual and presumed statistical models of the array data. Recently developed robust adaptive beamformers (RAB) combat the aforementioned mismatches at the cost of increased computational complexity by solving optimization problems with numerical search and/or iterative calculations. On the contrary, proposed in this paper is an analytically derived robust adaptive beamformer based on the closed-form, low-complexity estimation of the array data's interference-plus-noise covariance matrix (INCM). Instead of numerically/iteratively searching the actual steering vectors to maximize the Capon spectrum, the proposed estimator of the steering vector is analytically derived, by considering an inequality that assumes higher Capon spectrum intensity at the actual steering directions rather than the nominal ones. Consequently, the INCM is reconstructed using the closed-form estimates of interference steering vectors. Numerical simulations demonstrate both the computational efficiency and the robustness of the proposed beamformer against various types of mismatches. The proposed beamformer outperforms the conventional RAB's in terms of the output signal-to-interference-plus-noise-ratio (SINR), with order(s) of magnitude lower algorithm time, while not relying on any initial guess generally required in numerical or iterative searches.