A novel beamforming based on deconvolution for angular super-resolution

Deconvolved conventional beamforming (DCBF) has garnered significant attention in recent years due to its capacity to obtain high-resolution spatial spectra through deconvolving conventional beamforming (CBF) output with low computational complexity. However, the DCBF method suffers from the wide main lobe and low array gain (AG), resulting in angular resolution degradation under low signal-to-noise ratios (SNRs). To address these issues, a novel deconvolution super-resolution beamforming (DSRBF) algorithm is presented. Firstly, a super-resolution beamforming (SRBF) is constructed based on the split beamforming concept. The theoretical analysis indicates that SRBF exhibits a narrower main lobe and higher AG than CBF. Then, it is also proven that SRBF is approximately linear shift-invariant, and it can be further extended to deconvolution via constructing DSRBF, which provides a 3.4 dB AG improvement compared to DCBF. Thus, the proposed method can provide good performance on estimation and resolution. Finally, the simulation and experiment results are demonstrated to verify the theoretical analysis.