A metric to describe access point significance in location estimation

Abstract

Indoor localization is a key enabling technology for numerous location based services (LBS). A promising indoor localization technique is location fingerprinting (LF), having the major advantage of exploiting already existing radio infrastructures. LF can accurately estimate the user's location providing reliable RF fingerprints, that are unique and stable over time, are available. We propose a method to enhance the LF by exploiting the spatio-temporal characteristics of RF signals. The method quantifies an access point's (AP) significance for location estimation based on spatial uniqueness and temporal stability characteristics of its RF signals. Based on the proposed significance metric, APs contribute with different weights to the location estimation. By weighting the measurements, more reliable input data are provided to the localization algorithm which consequently results in improved LF performance.