A Theoretical Analysis on Site Survey in WiFi Fingerprint-based Localization

Xiaomeng Li,Baoqi Huang
DOI: https://doi.org/10.1109/ccdc52312.2021.9601713
2021-01-01
Abstract:The growing interest in indoor location-based services (LBSs) has stimulated many recent developments of indoor localization technologies. Due to the widespread deployment of Wireless Local Area Networks (WLAN), WiFi has become a most promising one. This paper discusses one of the key problems in WiFi fingerprint-based localization, namely how to conduct an efficient site survey. To this end, a probabilistic framework is first established to characterize the ability of distinguishing different fingerprints, and is then extended to multiple fingerprints in one dimensional scenarios. On these grounds, expected localization error is formulated by assuming the well known Lognormal distance path loss (LDPL) model, and moreover, the influence of the key site survey parameter, i.e. the interval between two neighboring reference points, is investigated. The theoretical analysis reveals that expected localization error increases with the interval between two reference points. Extensive simulations are conducted and confirm the effectiveness of this study.
What problem does this paper attempt to address?