Low-Complexity Adaptive Fourier Integral Method for Large-Aperture Horizontal Linear Arrays in Low-Frequency and Low-SNR Environments

The weak-signal detection poses a fundamental challenge in the passive sonar measurement systems, particularly under low-frequency and low signal-to-noise ratio (SNR) conditions. To address this challenge, this article proposes a low-complexity adaptive Fourier integral method (LCA-FIM), which introduces an innovative signal processing framework designed to enhance the measurement capabilities of large-aperture horizontal linear arrays (HLAs). The core innovation is an adaptive window selection strategy: multiple predesigned windows with different noise gains and sidelobe levels (SLLs) are applied to the Fourier integral method (FIM), generating several groups of weighted spatial spectra. Subsequently, an adaptive selection rule extracts the minimum output at each azimuth. This strategy enables LCA-FIM to effectively optimize the tradeoff between key performance metrics, achieving high noise gain, high angular resolution and low SLLs simultaneously. Numerical simulations demonstrate that LCA-FIM shows a better weak-signal detection performance under strong interference and low-SNR conditions, when compared with conventional beamformer, FIM, weighted FIM (WFIM), and diagonally loaded minimum variance distortionless response (MVDR) beamformer. Experimental validation using real ocean data further confirms the effectiveness of the proposed LCA-FIM method in clutter-rich, noisy, low-frequency and low-SNR environments. While validated in a passive sonar context, the proposed LCA-FIM is a general-purpose measurement method applicable to a wide range of array-based sensing system requiring weak-signal analysis.