Peer-to-Peer Determination of Proximity Using Wireless Network Data
PERCOMW '04 Proceedings of the Second IEEE Annual Conference on Pervasive Computing and Communications Workshops
In-Building Positioning: Modeling Location for Indoor World
DEXA '04 Proceedings of the Database and Expert Systems Applications, 15th International Workshop
Sensing and visualizing spatial relations of mobile devices
Proceedings of the 18th annual ACM symposium on User interface software and technology
An Adaptive Two-Phase Approach to WiFi Location Sensing
PERCOMW '06 Proceedings of the 4th annual IEEE international conference on Pervasive Computing and Communications Workshops
Fuzzy location and tracking on wireless networks
Proceedings of the 4th ACM international workshop on Mobility management and wireless access
GPS-based location extraction and presence management for mobile instant messenger
EUC'07 Proceedings of the 2007 international conference on Embedded and ubiquitous computing
WiFi localization system based on fuzzy logic to deal with signal variations
ETFA'09 Proceedings of the 14th IEEE international conference on Emerging technologies & factory automation
Discovering significant places from mobile phones: a mass market solution
MELT'09 Proceedings of the 2nd international conference on Mobile entity localization and tracking in GPS-less environments
WASP: an enhanced indoor locationing algorithm for a congested Wi-Fi environment
MELT'09 Proceedings of the 2nd international conference on Mobile entity localization and tracking in GPS-less environments
Information Sciences: an International Journal
| Hi-index | 0.00 |
This paper describes an algorithm to compute lists of people and devices that are physically nearby to a mobile user based on the analysis of signals from existing wireless networks. The system evaluates proximity by classifying the degree of similarity of the Wi-Fi scan data through a statistical Gaussian Mixture Model. It recognizes when the devices are in the same area, and, in this case, it distinguishes three proximity levels: High (e.g. same room), Medium (e.g. same floor) and Low (e.g. same building). The algorithm can be deployed on a remote server that receives Wi-Fi scanning data (including MAC addresses and signal strength) from mobile devices. The server estimates proximity by extracting a set of features from each received pair of Wi-Fi data, feeding them to the GMM model and selecting the category with greatest probability. The method presented in the paper does not require calibration and leverages on existing Wi-Fi signals, while obtaining a percentage of correct discrimination among three levels near to 90%.