Near-Field Source Localization via Symmetric Subarrays

Abstract

We propose a near-field source localization algorithm with one-dimensional (1-D) search via symmetric subarrays. By dividing the uniform linear array (ULA) into two symmetric subarrays, the steering vectors of the subarrays yield the 1-D (only bearing-related) property of rotational invariance in signal subspace, which allows for the bearing estimation using the generalized far-field ESPRIT. With the estimated bearing, the range estimation of each source is consequently obtained by defining the 1-D MUSIC spectrum. This algorithm transforms the two-dimensional search involved in the parameter estimation to a 1-D search, and it does not require high-order statistics computation in contrast with the traditional near-field high-order ESPRIT algorithm