Robust adaptive beamforming using an iterative FFT algorithm

• Proposed a robust adaptive beamformer with constraints on the beamwidth and the peak sidelobe level.
• This robust adaptive beamforming problem is transformed into a power pattern synthesis problem.
• A modified iterative FFT based method is proposed to synthesize this power pattern.
• This method is faster than the state-of-the-art algorithms, and is more robust against moving target signals and steering vector errors.

Adaptive beamformers will degrade in the presence of model mismatch. Because a wider beamwidth has higher ability against steering vector errors, and lower sidelobe levels can improve the robustness against fast moving interferences, in this work an iterative fast Fourier transform (FFT) based adaptive beamformer is proposed with constraints on beamwidth and peak sidelobe level. The adaptive beamforming is transformed to a weighted pattern synthesis problem. This weighted pattern is a product of the array pattern and a weighting function. Because the weighting function has shape peaks at the direction of interferences, it will have nulls in the array pattern at the directions of interferences by reducing the peak sidelobe level of this weighted pattern. A modified iterative FFT algorithm is proposed to synthesize this weighted pattern. Thanks to the efficiency of FFT, the nonconvex problem of power pattern synthesis can be solved efficiently. This method is demonstrated through several simulation examples. The results show the advantages of the proposed method in obtaining high output SINRs against moving target signals and steering vector errors.