Probabilistic Localization of Mobile Wireless LAN Client in Multistory Building Based on Sparse Bayesian Learning

Abstract

This paper describes a method for localizing wireless mobile clients in a multistory building using a public wireless Local Area Network (LAN) system. Physical location data on personal devices and mobile robots is important to information services and robot applications. Wireless mobile clients are localized in a multistory building using public wireless LAN access points placed, three-dimensionally in the building. Information on the floor number and client location is acquired probabilistically, with estimation providing a probabilistic model for localization based on sparse Bayesian learning. Results of experiments confirm the feasibility of our proposal.