Robust broadband adaptive beamforming with constant beamwidth using frequency difference technique

Constant beamwidth and robustness are two important features for broadband adaptive beamformers. In this paper, a robust broadband adaptive beamformer with a constant beamwidth for uniform linear arrays is proposed. Benefitting from the idea behind the frequency difference (FD) technique, the array observation in frequency bin f is first multiplied by the conjugate form of a reference array output in frequency bin f + f0, which is constructed with a given Gaussian random sequence from the looking direction. In this manner, a frequency-invariant property is obtained by mapping the frequency band of interest to a selected frequency bin f0. To improve the robustness of the proposed method, an iterative method with adaptive uncertainty in the steering vector (SV) is applied to the equivalent array output after the FD operation. The convergence can be accelerated by scaling the uncertainty level of the SV. Due to the FD technique, the proposed method adaptively filters the signal of interest in the FD domain. The effectiveness of the proposed method is verified by numerical simulation.& COPY; 2023 Elsevier Inc. All rights reserved.

Automated summary

In this paper, a robust broadband adaptive beamformer with a constant beamwidth for uniform linear arrays is proposed, which utilizes the frequency difference (FD) technique to achieve frequency-invariant property. An iterative method with adaptive uncertainty in the steering vector (SV) is applied to improve the robustness of the method. The effectiveness of the proposed method is verified by numerical simulation.