FinCCM: Fingerprint Crowdsourcing, Clustering and Matching for Indoor Subarea Localization

摘要

Fingerprinting based on received signal strength (RSS) is becoming a research focus in indoor localization. However, its time-consuming and labor-intensive site survey is a big hurdle for practical deployments. This letter proposes a novel indoor subarea localization scheme based on fingerprint passive crowdsourcing and unsupervised clustering, which first classifies unlabeled RSS measurements into several clusters and then relates clusters to indoor subareas to generate subarea fingerprints. In the online positioning phase, an observed fingerprint is located into the subarea with the least fingerprint difference. Our experimental results show that in typical indoor scenarios, the proposed scheme can achieve 95% subarea hitting rate to correctly locate a smartphone to its subarea.