Subspace Fitting Approach for Wideband Near-Field Localization

TL;DR

This paper introduces two subspace fitting methods for wideband near-field localization, effectively estimating distance and angle jointly or separately, with verified numerical performance.

Contribution

It proposes novel subspace fitting algorithms tailored for wideband near-field systems, including a joint estimation method and a complexity-reducing approximation.

Findings

Both methods accurately estimate target parameters.

The Fresnel approximation reduces computational complexity.

Numerical results confirm the effectiveness of the approaches.

Abstract

Two subspace fitting approaches are proposed for wideband near-field localization. Unlike in conventional far-field systems, where distance and angle can be estimated separately, spherical wave propagation in near-field systems couples these parameters. We therefore derive a frequency-domain near-field signal model for multi-target wideband systems and develop a subspace fitting-based MUSIC method that jointly estimates distance and angle. To reduce complexity, a Fresnel approximation MUSIC algorithm is further introduced to decouple the distance and angle parameters. Numerical results verify the effectiveness of both proposed approaches.