Broadband signals DOA estimation based on two-dimensional array manifold interpolation

Abstract

The array manifold interpolation (AMI) method can effectively address the need for angle pre-estimation in focusing transformation methods for broadband signal direction of arrival (DOA) estimation. However, traditional AMI methods are constrained by array geometry and signal incident direction, limiting their applicability, especially for volumetric arrays. To tackle these challenges, this paper considers the scenario where broadband signals impinge on arrays of arbitrary geometries from arbitrary directions. By utilizing the Jacobi-Anger expansion and Legendre series, the array manifold matrix is decomposed to obtain array manifold interpolation matrix, and a focusing transformation matrix suitable for arbitrary array configurations is derived based on these results. Compared to conventional methods, this approach extends the applicability of AMI methods to volumetric arrays, achieving higher DOA estimation accuracy and enhanced angular resolution in low signal-to-noise ratio (SNR) environments, while reducing computational complexity. Simulations and real data processing results verify the effectiveness of the proposed algorithm.