ScOFi: Schematic assisted optimum fingerprinting for Wi-Fi indoor localization using peer hand-shake

Abstract

Received signal strength fingerprints based on Wi-Fi spectrum have been widely adopted in the recent years for indoor localization purposes due to cost-effectiveness and availability. However, until the peer hand-shake (PHS), existing work had not constrained the schematic dimension of the target area, which could dramatically reduce the localization error. As the demand for sensors everywhere schemes for 5G networks keeps on booming, effective signal propagation characterization with in indoors is very essential for Internet of things (IoT) indoor localization and navigation applications. We review, extend the validation of the PHS technique that leverages the schematic dimensions of the target area within the total indoor environment to construct, auto-dynamically transform and update fingerprint in complex indoor environments. Extensive experimental validation has been carried out in two scenarios; Scenario 1 categorizes lobby area while Scenario 2 categorizes corridor areas. We analyze the accuracy performance using Nearest Neighbor (NN) and the KNN algorithms. Experimental results show robustness of the PHS, achieving lower average localization error in diverse indoor dimensionalities than comparisons.