Deconvolved Beamforming Based on Fractional Fourier Transform

Abstract

To estimate the direction of arrival (DOA) of broadband linear frequency modulation (LFM) signals in complex underwater environment, where hydrophone available is few and signal-to-noise ratio (SNR) is low, deconvolved fractional Fourier domain beamforming (DFrFB) is proposed. Fractional Fourier domain delay-and-sum beamforming (FrFB) is generally used to deal with this issue, however, it produces a wide beamwidth and high sidelobes in spatial spectrum. In this paper, deconvolution is introduced in fractional Fourier domain to decrease the beamwidth and sidelobes, based on the shift invariance of FrFB beam pattern. Simulation and experiment results show that, the beamwidth of DFrFB is 80% narrower than that of FrFB and sidelobe levels of DFrFB are 10dB lower than that of FrFB. The proposed DFrFB can yield a narrow beam and low sidelobes with great robustness on the condition of few hydrophones and low SNR.

Keywords: Deconvolved beamforming, fractional Fourier transform, direction of arrival estimation