For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Content-Based Sensor Search with a Matching Estimation Mechanism
Puning ZHANG Yuan-an LIU Fan WU Wenhao FAN Bihua TANG
IEICE TRANSACTIONS on Communications
Publication Date: 2016/09/01
Online ISSN: 1745-1345
Type of Manuscript: Special Section PAPER (Special Section on Integration Technologies of Ambient Intelligence and Sensor Networks)
Internet of Things, content-based, sensor search, matching estimation,
Full Text: PDF(2.1MB)>>
The booming developments in embedded sensor technique, wireless communication technology, and information processing theory contribute to the emergence of Internet of Things (IoT), which aims at perceiving and connecting the physical world. In recent years, a growing number of Internet-connected sensors have published their real-time state about the real-world objects on the Internet, which makes the content-based sensor search a promising service in the Internet of Things (IoT). However, classical search engines focus on searching for static or slowly varying data, rather than object-attached sensors. Besides, the existing sensor search systems fail to support the search mode based on a given measurement range. Furthermore, accessing all available sensors to find sought targets would result in tremendous communication overhead. Thus an accurate matching estimation mechanism is proposed to support the search mode based on a given search range and improve the efficiency and applicability of existing sensor search systems. A time-dependent periodical prediction method is presented to periodically estimate the sensor output, which combines with the during the period feedback prediction method that can fully exploit the verification information for enhancing the prediction precision of sensor reading to efficiently serve the needs of sensor search service. Simulation results demonstrate that our prediction methods can achieve high accuracy and our matching estimation mechanism can dramatically reduce the communication overhead of sensor search system.