Radio Map Building with IEEE 802.15.4 for Indoor Localization Applications

Abstract

Automated radio map building for Radio Signal Strength (RSS) based fingerprinting localization is implemented in this paper. Fingerprinting technique is composed of an offline phase for collecting geotagged fingerprint data called a radio map and an online phase for location estimation by comparing an unknown object’s location with the radio map. Typically, radio map is created manually by moving and pin-pointing a location while scanning a network for RSS values within a pre-specified pattern. This cumbersome method is very time consuming and results in inaccuracies in the radio map. This paper presents an automated system to build a radio map of indoor environment by using a self-directed car. The car can move, measure 2D distances, collect RSS values and forward the data to a server via a gateway which calculates the car location coordinates. To test and verify the system, an experiment is conducted in our departmental office spaces. The coordinate estimation accuracy and RSS mapping consistency produced by the proposed system make it possible to build a more accurate and realistic radio map for indoor environments in an automated fashion with decreased possibility of human errors.