Seamless switching between GNSS and WLAN based indoor positioning system for ubiquitous positioning

Abstract

Ubiquitous positioning technology is a term coined to address the requirements of specifying the “whereness” completely. The technology comprises the indoor and outdoor positioning systems. The Global Navigation Satellite System (GNSS) based systems like GPS, GLONASS, Galileo and QZSS are some of the systems that provide outdoor positioning. While Indoor Positioning Systems (IPSs) are undergoing rapid development, the systems can be supplied using short range wireless technologies such as Wi-Fi, Bluetooth, RFID, and Infrared. Wi-Fi based positioning systems are being particularly intensely researched because of their ubiquitous presence. In this paper, position determination in a 3D indoor environment is explored using the Received Signal Strength Indication (RSSI) as an input to multi-layer feed forward-back propagation artificial neural networks (ANNs), which are then integrated with GNSS to create a Ubiquitous Positioning System (UPS). The paired ANN model, an enhanced data collection method, and the architecture of the UPS are also presented in this paper, where the UPS was thoroughly investigated on a real Wi-Fi network. The result showed that the paired ANN attains a 30 % increase in accuracy compared to a single ANN, while the UPS shows a mean distance error of 3.5 m.