Low-complexity robust beamforming via orthogonal polynomial fitting and dual-phase-shifter structure

Abstract

In adaptive beamforming-based anti-interference systems, dynamic interferences or platform perturbations can cause misalignment between interference signals and pre-configured spatial nulls, leading to significant degradation in anti-interference performance due to reduced suppression capability of the beamforming algorithm. To prevent such misalignment, broadening the width of the spatial nulls is a feasible solution. However, conventional methods suffer from high computational complexity due to the integration and inversion operations involving the covariance matrix. This paper proposes a robust adaptive beamforming algorithm with reduced complexity by reconstructing the covariance matrix. We employ orthogonal polynomial fitting to approximate the steering vectors associated with the broadened nulling regions. Hence, the interference subspace is directly obtained and the robustness of the interference suppression is enhanced while maintaining low complexity. Additionally, the algorithm is implemented based on the double-phase shifters (DPS) structure to control the antenna weights and phases. Compared to the traditional attenuator and single-phase shifter (ATT-SPS) structure, the DPS structure represents the weight vector as the sum of two constant modulus vectors. Simulation results indicate that the DPS structure performs better in beamforming at lower quantization bits, achieving performance closer to that of an unquantized system.