Research Article
An Accurate Passive RFID Indoor Localization System Based on Sense-a-Tag and Zoning Algorithm
@INPROCEEDINGS{10.1007/978-3-319-51204-4_22, author={Majed Rostamian and Jing Wang and Miodrag Bolić}, title={An Accurate Passive RFID Indoor Localization System Based on Sense-a-Tag and Zoning Algorithm}, proceedings={Ad Hoc Networks. 8th International Conference, ADHOCNETS 2016, Ottawa, Canada, September 26-27, 2016, Revised Selected Papers}, proceedings_a={ADHOCNETS}, year={2017}, month={4}, keywords={Internet of Things Ultra High Frequency (UHF) Radio Frequency Identification (RFID) Weighted centroid Zoning algorithm}, doi={10.1007/978-3-319-51204-4_22} }
- Majed Rostamian
Jing Wang
Miodrag Bolić
Year: 2017
An Accurate Passive RFID Indoor Localization System Based on Sense-a-Tag and Zoning Algorithm
ADHOCNETS
Springer
DOI: 10.1007/978-3-319-51204-4_22
Abstract
Localization and tracking of objects (e.g. objects or people) in indoor environment will facilitate many location dependent or context-aware applications. Localization of passive ultra high frequency (UHF) radio-frequency identification (RFID) tags attached to objects or people is of special interest because of the low cost of the tags and backscatter communication that is power efficient. An augmented RFID system for localization based on a new tag called Sense-a-Tag (ST) that communicates with the RFID reader as a passive tags and can detect and record communication of other passive tags in its proximity was introduced several years ago. In ST-based localization system, a large set of passive landmark tags are placed at the known locations. The system localizes ST based on the aggregation of binary detection measurements according to localization algorithm, such as weighted centroid localization (WCL). However, the aforementioned method is easily affected by the outlier detection of distant landmark tags by ST. To improve localization accuracy, this paper propose to iteratively refine the interrogation area of the reader so that it includes only the most relevant landmark tags. The performance of the proposed method is demonstrated by extensive computer simulation and realistic experiments.