부산대학교 도서관

The Conventional Beamforming (CBF) Method has been widely used for underwater target detection. However, the performance and resolution of the CBF are insufficient in low signal-to-noise ratio (SNR) cases. To circumvent such a situation, this article proposes a new short angle weight (SAW) algorithm to improve the performance and resolution for target detection. Compared with the CBF, the proposed approach only uses the main lobe of the azimuth spectrum, while cutting off the side lobes due to redundancy. For obtaining the spatial power distribution and estimating the target azimuth, a weighted dictionary matrix is constructed to deconvolute the azimuth spectrum. Since the resolution of the azimuth spectrum is determined by the rate at which it changes, by properly selecting the size of the support set of its Fourier-transformed function, the resolution of the azimuth spectrum can apparently be enhanced. Moreover, the performance of the proposed SAW method is validated by pool trials and lake trials. The corresponding results show that the proposed algorithm provides much better resolution than other representative methods, at least for all the tested instances.