Map-Aware Models for Indoor Wireless Localization Systems: An Experimental Study

TL;DR

This paper develops and experimentally validates a unified map-aware statistical model for indoor wireless localization, demonstrating significant accuracy improvements over map-unaware methods, especially with poor measurement quality.

Contribution

It introduces a novel unified statistical model for map-aware indoor localization based on TOA and RSS measurements, validated through experiments.

Findings

Map-aware modeling improves accuracy by up to 110% in poor measurement scenarios.

The proposed model outperforms map-unaware counterparts in experimental tests.

Experimental validation confirms the effectiveness of the unified statistical approach.

Abstract

The accuracy of indoor wireless localization systems can be substantially enhanced by map-awareness, i.e., by the knowledge of the map of the environment in which localization signals are acquired. In fact, this knowledge can be exploited to cancel out, at least to some extent, the signal degradation due to propagation through physical obstructions, i.e., to the so called non-line-of-sight bias. This result can be achieved by developing novel localization techniques that rely on proper map-aware statistical modelling of the measurements they process. In this manuscript a unified statistical model for the measurements acquired in map-aware localization systems based on time-of-arrival and received signal strength techniques is developed and its experimental validation is illustrated. Finally, the accuracy of the proposed map-aware model is assessed and compared with that offered by its map-unaware counterparts. Our numerical results show that, when the quality of acquired measurements is poor, map-aware modelling can enhance localization accuracy by up to 110% in certain scenarios.