The Community for Technology Leaders
Green Image
Issue No. 11 - Nov. (2012 vol. 23)
ISSN: 1045-9219
pp: 2138-2149
Yunhao Liu , Tsinghua University and Hong Kong University of Science and Technology, Hong Kong
Yiyang Zhao , Hong Kong University of Science and Technology, Hong Kong
Lei Chen , Tsinghua University, China
Jian Pei , Simon Fraser University, Burnaby
Jinsong Han , Xi'an Jiaotong University, China
Activity monitoring, a crucial task in many applications, is often conducted expensively using video cameras. Effectively monitoring a large field by analyzing images from multiple cameras remains a challenging issue. Other approaches generally require the tracking objects to attach special devices, which are infeasible in many scenarios. To address the issue, we propose to use RF tag arrays for activity monitoring, where data mining techniques play a critical role. The RFID technology provides an economically attractive solution due to the low cost of RF tags and readers. Another novelty of this design is that the tracking objects do not need to be equipped with any RF transmitters or receivers. By developing a practical fault-tolerant method, we offset the noise of RF tag data and mine frequent trajectory patterns as models of regular activities. Our empirical study using real RFID systems and data sets verifies the feasibility and the effectiveness of this design.
Radio frequency, Trajectory, Monitoring, Radiofrequency identification, Data mining, Cameras, Interference, trajectory, Active RFID, mining

L. Chen, Y. Zhao, Y. Liu, J. Pei and J. Han, "Mining Frequent Trajectory Patterns for Activity Monitoring Using Radio Frequency Tag Arrays," in IEEE Transactions on Parallel & Distributed Systems, vol. 23, no. , pp. 2138-2149, 2012.
178 ms
(Ver 3.3 (11022016))