Neighbor-Assisted Localization for Massive MIMO 5G Systems

Abstract

In this paper, we propose a multi-stage localization technique that enables to determine the position of a Target User Equipment (T-UE) with harsh channel conditions through the assistance of neighboring User Equipments (UEs). The proposed approach allows a reduced-complexity localization by minimizing the search space through multistage treatment. A neighbor discovery processing is first performed to identify the two nearest neighboring UEs among a set of UEs in the vicinity. The so selected neighbors are used as anchors (A-UE). Then, based on the signal strength, the distances separating the two A-UEs to the T-UE are determined. The distance estimates are used to plot two potential solutions of the Angle of Arrival (AoA) of T-UE. To distinguish the correct AoA estimate, oriented beamforming is performed around short arcs centered on the two AoA candidates. The user's position is then deduced based on the estimates of the AoA and the distance to the A-UEs. Our approach exploits the capabilities of neighbor discovery and oriented beamforming to provide an accurate position estimate for UEs experiencing harsh channel conditions that render direct localization at the base station difficult.