Bootstrapped Learning of WiFi Access Point in Hybrid Positioning System

Jing Yu, Thandapani Venkataramani, Xing Zhao, Zhike Jia

Abstract:	In a GNSS/Wi-Fi/Sensors integrated navigation system, Wi-Fi is responsible for providing absolute position updates indoors, at anytime. The triangulation method is generally preferred in Wi-Fi positioning for covering large-scale ranges. This method is reliable; however, the main implementation challenge is dealing with the lack of the geographic information for the local Wi-Fi access points. This paper explores a bootstrapped algorithm that can derive the AP coordinate by itself; then the derived knowledge can be maintained locally on the mobile device for positioning. If no prior Wi-Fi knowledge is available, the engine selects GNSS/sensors positioning solution as known surveying points so that the coordinates of the AP can be derived in a real-time. Extensive field tests have been performed using Marvell’s leading smartphone platform to verify the algorithms. The system can derive around 80% of the available Wi-Fi access points within an uncertainty of 25 meters. As a result, GNSS/MEMS/Wi-Fi hybrid positioning accuracies of 10 meters can be obtained for over 95% of time in indoor walking tests. The bootstrapped method of writing the Wi-Fi AP database alleviates the dependency for apriori infrastructure knowledge, thus it promotes the practical deployment of a self-contained indoor positioning product for the consumer market.
Published in:	Proceedings of the 25th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2012) September 17 - 21, 2012 Nashville Convention Center, Nashville, Tennessee Nashville, TN
Pages:	960 - 966
Cite this article:	Yu, Jing, Venkataramani, Thandapani, Zhao, Xing, Jia, Zhike, "Bootstrapped Learning of WiFi Access Point in Hybrid Positioning System," Proceedings of the 25th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2012), Nashville, TN, September 2012, pp. 960-966.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In