A Robust Beamforming Algorithm for Sparse Array Based on Atomic Norm Minimization

Aiming at the performance degradation of the beamforming algorithm when some mismatches are present in the signal model. A robust beamforming algorithm based on atomic norm minimization (ANM) is proposed to improve the beamforming performance of sparse array during the mismatch of signal model. The proposed algorithm is used to construct an ANM-based noise reduction model and transform it into an equivalent semi-definite programming problem according to the covariance matrix structure of sparse array. Meanwhile, the dual problem of this model is derived to improve the computational efficiency, and the received data and covariance matrix of the array after noise reduction are obtained. The spatial spectrum is proved to be unambiguous based on the structural properties of a coprime array, and the directions of arrival of the incident signals are obtained directly by using the multiple signal classification algorithm for the resulting covariance matrix. The received data of a uniform linear array with the same aperture as the coprime array is obtained using the virtual filling technique, and the array output is ultimately obtained. Simulated results verify the feasibility and accuracy of the proposed algorithm, which improves the output signal to interference plus noise ratio by at least 1. 5 dB compared to the other tested algorithms. © 2024 China Ordnance Industry Corporation. All rights reserved.