Landmark-Based Indoor Positioning: a Non-Intrusive and Cost-Effective Approach

Abstract

This paper proposes a method for identifying indoor landmarks that can serve as fingerprints in indoor environments. The reason for this proposal is that every building has several distinctive locations. These signatures will be used as positional markers if they can be found. Landmarks such as Wi-Fi, entrance, windows, and elevators are recognizable. The proposed method detects these landmarks as follows: Wi-Fi Received Signal Strength (RSS) is used to detect the Wi-Fi landmark. The difference between GPS signals and Wi-Fi RSS values is used to locate the entrance landmark. GPS signals, though not available indoors, can be detected near windows and used to locate windows. Elevators are identified using two criteria: user movement behavior and loss of Wi-Fi RSS. The accuracy of detecting Wi-Fi, entrance, and window landmarks is 2 m, while the accuracy of detecting the elevator landmark is 1 m.