Share Email Print

Proceedings Paper

Implementation of a compressive sampling scheme for wireless sensors to achieve energy efficiency in a structural health monitoring system
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Wireless sensors have emerged to offer low-cost sensors with impressive functionality (e.g., data acquisition, computing, and communication) and modular installations. Such advantages enable higher nodal densities than tethered systems resulting in increased spatial resolution of the monitoring system. However, high nodal density comes at a cost as huge amounts of data are generated, weighing heavy on power sources, transmission bandwidth, and data management requirements, often making data compression necessary. The traditional compression paradigm consists of high rate (>Nyquist) uniform sampling and storage of the entire target signal followed by some desired compression scheme prior to transmission. The recently proposed compressed sensing (CS) framework combines the acquisition and compression stage together, thus removing the need for storage and operation of the full target signal prior to transmission. The effectiveness of the CS approach hinges on the presence of a sparse representation of the target signal in a known basis, similarly exploited by several traditional compressive sensing applications today (e.g., imaging, MRI). Field implementations of CS schemes in wireless SHM systems have been challenging due to the lack of commercially available sensing units capable of sampling methods (e.g., random) consistent with the compressed sensing framework, often moving evaluation of CS techniques to simulation and post-processing. The research presented here describes implementation of a CS sampling scheme to the Narada wireless sensing node and the energy efficiencies observed in the deployed sensors. Of interest in this study is the compressibility of acceleration response signals collected from a multi-girder steel-concrete composite bridge. The study shows the benefit of CS in reducing data requirements while ensuring data analysis on compressed data remain accurate.

Paper Details

Date Published: 11 April 2013
PDF: 11 pages
Proc. SPIE 8694, Nondestructive Characterization for Composite Materials, Aerospace Engineering, Civil Infrastructure, and Homeland Security 2013, 86941L (11 April 2013); doi: 10.1117/12.2010128
Show Author Affiliations
Sean M. O'Connor, Univ. of Michigan (United States)
Jerome P. Lynch, Univ. of Michigan (United States)
Anna C. Gilbert, Univ. of Michigan (United States)

Published in SPIE Proceedings Vol. 8694:
Nondestructive Characterization for Composite Materials, Aerospace Engineering, Civil Infrastructure, and Homeland Security 2013
Tzu Yang Yu, Editor(s)

© SPIE. Terms of Use
Back to Top