Graph-based Efficient WiFi Fingerprint Training Using Un-supervised Learning

Bo Zhao, Ling Pei, Changqing Xu, Li Gu

Peer Reviewed

Abstract:	WiFi localization systems need an offline phase to generate a fingerprint database, which costs a tremendous amount of workload. To overcome this problem, we develop a new graph-based fingerprint training method using un-supervised learning. In the proposed method, graph structure of indoor environments is firstly built. Subsequently, sampling path from the Starting Point (SP) to the Turning Point (SP) is computed with the method base on direction matching. Finally, the samples on the path are obtained by linear interpolating. Fingerprint database can be derived from these steps. Then the database can be improved for a higher positioning accuracy and a wider scope of application by Received Signal Strength (RSS) modeling and samples clustering. Since the procedure of generating fingerprint database can be completed simultaneously with daily activities, it is unessential to spend specific labor for fingerprint training, which significantly decreases the workload of fingerprint database building, and guarantee positioning accuracy at the same time.
Published in:	Proceedings of the 28th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2015) September 14 - 18, 2015 Tampa Convention Center Tampa, Florida
Pages:	2301 - 2310
Cite this article:	Zhao, Bo, Pei, Ling, Xu, Changqing, Gu, Li, "Graph-based Efficient WiFi Fingerprint Training Using Un-supervised Learning," Proceedings of the 28th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2015), Tampa, Florida, September 2015, pp. 2301-2310.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In