A Spatially-interpolated beamformer using a parallel structure with sparse linear arrays

Abstract

A spatially-interpolated (SI) beamformer is proposed in this paper to enhance narrowband plane waves received by a sparse linear antenna array. Similar to a conventional SI filter, the proposed beamformer employs two filters: an SI prototype filter and a masking filter. However, the proposed SI beamformer employs a parallel structure compared to a cascade structure of conventional SI filters leading to a sparse antenna array. Both the prototype filter and the masking filter are designed as optimal filters in the minimax sense. Furthermore, we employ a bell-shaped passband for the ideal magnitude response for the design compared to a conventional ideal frequency response that employs 1 and 0 as the passband and the stopband gains with a finite transition band. Simulation results confirm that the proposed SI beamformer provides approximately 37% reduction in computational complexity and a similar reduction in the number of antennas required in the associated antenna array compared to other minimax designs, however, without spatial interpolation. Moreover, the proposed beamformer achieves around 20 dB improvement in the peak sidelobe level compared to an SI beamformer designed with the conventional ideal magnitude response.