Super-Resolution DOA Estimation for Wideband Signals using Arbitrary Linear Arrays without Focusing Matrices

TL;DR

This paper introduces a novel wideband DOA estimation method for arbitrary linear arrays that uses convex optimization without focusing matrices, achieving high accuracy and robustness without initial estimates.

Contribution

It proposes a convex optimization-based wideband DOA estimation technique that eliminates the need for focusing matrices and initial estimates, applicable to any linear array.

Findings

Outperforms existing methods in accuracy and noise robustness

Effective for closely spaced sources

No need for focusing matrices or initial DOA guesses

Abstract

We focus on developing an effective Direction Of Arrival (DOA) estimation method for wideband sources based on the gridless sparse concept. Previous coherent methods have been designed by dividing wideband frequencies into a few subbands which are transferred to a reference subband using focusing matrices. In this work, as opposed to the previous techniques, we propose a convex optimization problem that leads to an accurate wideband DOA estimation method with no need for any focusing matrix. Moreover, in this method, no initial DOA estimates are required and it can be used for any arbitrary linear arrays. Numerical simulations show that in comparison to some well-known techniques, the proposed method generates outstanding accuracy and better robustness to noise. The effectiveness of the method is also verified in presence of close adjacent sources.