Research Article
A Middleware Solution for Optimal Sensor Management of IoT Applications on LTE Devices
@INPROCEEDINGS{10.1007/978-3-319-60717-7_28, author={Satyajit Padhy and Hsin-Yu Chang and Ting-Fang Hou and Jerry Chou and Chung-Ta King and Cheng-Hsin Hsu}, title={A Middleware Solution for Optimal Sensor Management of IoT Applications on LTE Devices}, proceedings={Quality, Reliability, Security and Robustness in Heterogeneous Networks. 12th International Conference, QShine 2016, Seoul, Korea, July 7--8, 2016, Proceedings}, proceedings_a={QSHINE}, year={2017}, month={8}, keywords={IoT applications Sensor management LTE device Context-aware}, doi={10.1007/978-3-319-60717-7_28} }- Satyajit Padhy
Hsin-Yu Chang
Ting-Fang Hou
Jerry Chou
Chung-Ta King
Cheng-Hsin Hsu
Year: 2017
A Middleware Solution for Optimal Sensor Management of IoT Applications on LTE Devices
QSHINE
Springer
DOI: 10.1007/978-3-319-60717-7_28
Abstract
After many devices that have adopted LTE technology, it is optimistic to presume that 5G technology will have to address the huge traffic of data and volume of heterogeneous devices in future. Existing context-aware Internet of Things (IoT) applications directly control sensors on LTE devices in an uncoordinated and non-optimized manner, which leads to redundant sensor activations and energy wastage on resource-constrained IoT devices. Optimal and coordinated sensor usage dictates a comprehensive middleware solution to bring together the information from all IoT applications/sensors and intelligently select the best set of sensors to activate. In this paper, we design, implement, and evaluate a sensor management middleware for LTE devices that controls the tradeoff between energy consumption of sensors and accuracy of inferred contexts. The core task of this middleware is to minimize total energy consumption while making sure that the accuracy requested by IoT applications are met. Trace-driven simulations are conducted to demonstrate the merits of the proposed middleware and algorithms. The simulation results indicate that the proposed algorithms clearly outperform the current solution.
