Least Squares-Based Near-Field Source Localisation With Enhanced Range Estimation

Abstract

This letter proposes a least squares-based method for a single near-field source localisation by jointly estimating bearing angle and range using a uniform linear array. To provide initial estimates in one step, the proposed method uses least squares formulation developed based on the observation that adjacent sensor pairs exhibit far-field characteristics while the overall array has near-field properties. To further enhance the performance of range estimation, we re-estimate the range parameters by utilising geometric relationships. Consequently, the proposed method can provide improved localisation performance with reduced computational complexity. Simulation results demonstrate the superiority of the proposed method including the scheme of enhanced range estimation.