Coprime Array Interpolation for Direction of Arrival Estimation Based on Atomic Norm Minimization

Abstract

As an emerging non-uniform linear array, the coprime array (CA) can achieve more degrees of freedom (DOF) than the traditional uniform linear array (ULA). However, when using difference coarray, the holes in the virtual symmetric nonuniform linear array (VSNLA) prevent us from fully exploiting the whole virtual array aperture in CA. In this paper, we present a novel CA interpolation algorithm based on atomic norm minimization (ANM) to estimate the direction of arrival (DOA). The proposed method maps the observation vector of VSNLA to that of the virtual filled ULA with the same aperture size via ANM, and achieves better performance compared with the previous interpolation works such as the nuclear norm minimization (NNM) and the iterative initial DOA estimation based array interpolation method (ICA-AI). Simulation results validate the effectiveness of the proposed algorithm.