Indoor Localization Based on Weighted Surfacing from Crowdsourced Samples

Abstract

Fingerprinting-based indoor localization suffers from its time-consuming and labor-intensive site survey. As a promising solution, sample crowdsourcing has been recently promoted to exploit casually collected samples for building offline fingerprint database. However, crowdsourced samples may be annotated with erroneous locations, which raises a serious question about whether they are reliable for database construction. In this paper, we propose a cross-domain cluster intersection algorithm to weight each sample reliability. We then select those samples with higher weight to construct radio propagation surfaces by fitting polynomial functions. Furthermore, we employ an entropy-like measure to weight constructed surfaces for quantifying their different subarea consistencies and location discriminations in online positioning. Field measurements and experiments show that the proposed scheme can achieve high localization accuracy by well dealing with the sample annotation error and nonuniform density challenges.

Crowdsourcing Indoor Positioning by Light-Weight Automatic Fingerprint Updating via Ensemble Learning

“…Crowdsensing approaches [25]- [27] have recently been adopted to position and automatically update radio maps for leveraging crowdsourcing inertial sensor data. Relying on the crowdsourcing user traces constructed by inertial sensors, trace matching algorithms are designed to establish original user walking paths and generate radio maps [20], [28], [29], [30] and [31].…”

“…The deployment of additional receivers and reference points is expensive and non-scalable. Lin et al [27] propose a cross-domain cluster intersection algorithm to weight each sample reliability and construct radio propagation surfaces by polynomial functions matching. However, this work assumes that each crowdsourced sample has been annotated with a location, which is hard to get and the predicted location could be incorrect if altered APs exist.…”