Communications of the ACM
Crowdsourcing a Practical Indoor Gps
Researchers at the Korea Advanced Institute of Science and Technology have tapped smartphones to significantly reduce the cost of constructing an indoor localization system while maintaining high accuracy.
Credit: Asian Scientist
Researchers at the Korea Advanced Institute of Science and Technology are using crowdsourced Wi-Fi fingerprints from smartphones to inexpensively build a highly accurate indoor global-positioning system.
The research is based on an earlier method for automatically labeling resting-space locations from signals collected in various contexts such as homes, shops, and offices using people's home or office address information. The technique enables automatic labeling of transient-space locations without any additional location data.
The team analyzed indoor space usage to accurately glean and label the location information of the Wi-Fi fingerprints, and from there they transitioned to technology that classified indoor spaces as places used for stationary tasks and spaces used to reach such locations while applying separate algorithms to optimally and automatically gather location-labeling data.
Testing found the technology is accurate up to three or four meters with sufficient training data.
From Asian Scientist
View Full Article
Abstracts Copyright © 2017 Information Inc., Bethesda, Maryland, USA
No entries found