Abhishek B. Sharma
Ramesh Govindan
Abstract
Various sensor network measurement studies have reported instances of transient faults in sensor readings. In this work, we seek to answer a simple question: How often are such faults observed in real deployments? We focus on three types of transient faults, caused by faulty sensor readings that appear abnormal. To understand the prevalence of such faults, we first explore and characterize four qualitatively different classes of fault detection methods. Rule-based methods leverage domain knowledge to develop heuristic rules for detecting and identifying faults. Estimation methods predict “normal” sensor behavior by leveraging sensor correlations, flagging anomalous sensor readings as faults. Time-series-analysis-based methods start with an a priori model for sensor readings. A sensor measurement is compared against its predicted value computed using time series forecasting to determine if it is faulty. Learning-based methods infer a model for the “normal” sensor readings using training data, and then statistically detect and identify classes of faults.
We find that these four classes of methods sit at different points on the accuracy/robustness spectrum. Rule-based methods can be highly accurate, but their accuracy depends critically on the choice of parameters. Learning methods can be cumbersome to train, but can accurately detect and classify faults. Estimation methods are accurate, but cannot classify faults. Time-series-analysis-based methods are more effective for detecting short duration faults than long duration ones, and incur more false positives than the other methods. We apply these techniques to four real-world sensor datasets and find that the prevalence of faults as well as their type varies with datasets. All four methods are qualitatively consistent in identifying sensor faults, lending credence to our observations. Our work is a first step towards automated online fault detection and classification.
- Balzano, L. and Nowak, R. 2007. Blind calibration in sensor networks. In Proceedings of the Intenational Conference on Information Processing in Sensor Networks (IPSN). Google Scholar
Digital Library
- Bengio, Y. and Frasconi, P. 1995. An input output HMM architecture. In Proceedings of the Neural Information Processing Systems Conference (NIPS).Google Scholar
- Box, G. E. P., Jenkins, G. M., and Reinsen, G. C. 1994. Time Series Analysis: Forecasting and Control, 3rd Ed. Prentice Hall. Google ScholarDigital Library
- Bychkovskiy, V., Megerian, S., Estrin, D., and Potkonjak, M. 2003. A collaborative approach to in-place sensor calibration. In Proceedings of the 2nd Intenational Workshop on Information Processing in Sensor Networks (IPSN). Google ScholarDigital Library
- Chatfield, C. 2000. Time Series Forecasting. Chapman and Hall/CRC Press.Google Scholar
- Elnahrawy, E. and Nath, B. 2003. cleaning and querying noisy sensors. In Proceedings of the ACM International Workshop on Wireless Sensor Networks and Applications (WSNA). Google ScholarDigital Library
- Gupchup, J., Sharma, A., Terzis, A., Burns, R., and Szalay, A. 2008. The perils of detecting measurement faults in environmental monitoring networks. In Proceedings of the ProSense Special Session and International Workshop on Wireless Sensor Network Deployments (WiDeploy), held at DCOSS.Google Scholar
- INTEL. 2004. The Intel Lab Data. Dataset available at http://berkeley.intel-research.net/labdata/.Google Scholar
- Jeffery, S. R., Alonso, G., Franklin, M. J., Hong, W., and Widom, J. 2006. Declarative support for sensor data cleaning. In Proceedings of the International Conference on Pervasive Computing. Google ScholarDigital Library
- Kailath, T., Ed. 1977. Linear Least-Squares Estimation. Hutchison & Ross, Stroudsburg, PA.Google Scholar
- Khoussainova, N., Balazinska, M., and Suciu, D. 2006. Towards correcting input data errors probabilistically using integrity constraints. In Proceedings of the ACM Workshop on Data Engineering and Mobile Access (MobiDE). Google ScholarDigital Library
- Koushanfar, F., Potkonjak, M., and Sangiovammi-Vincentelli, A. 2003. On-Line fault detection of sensor measurements. IEEE Sensors J.Google Scholar
- Mainwaring, A., Polastre, J., Szewczyk, R., and Anderson, D. C. J. 2002. Wireless sensor networks for habitat monitoring. In the ACM International Workshop on Wireless Sensor Networks and Applications (WSNA). Google ScholarDigital Library
- MATLAB. http://www.mathworks.com/.Google Scholar
- NAMOS. 2005. NAMOS: Networked aquatic microbial observing system. Dataset available at http://robotics.usc.edu/~namos/data/jr_oct/web/.Google Scholar
- NAMOS. 2006. NAMOS: Networked aquatic microbial observing system. Dataset available at http://robotics.usc.edu/~namos/data/jr_aug_06/.Google Scholar
- Ni, K., Ramanathan, N., Chehade, M., Balzano, L., Nair, S., Zahedi, S., Pottie, G., Hansen, M., and Srivastava., M. 2008. Sensor network data fault types. Trans. Sensor Netw. To appear. Google ScholarDigital Library
- Rabiner, L. 1989. A tutorial on hidden Markov models and selected applications in speech recognition. Proc. IEEE 77, 2, 257--286.Google ScholarCross Ref
- Ramanathan, N., Balzano, L., Burt, M., Estrin, D., Kohler, E., Harmon, T., Harvey, C., Jay, J., Rothenberg, S., and Srivastava, M. 2006a. Rapid deployment with confidence: Calibration and fault detection in environmental sensor networks. Tech. rep. 62, CENS.Google Scholar
- Ramanathan, N., Schoellhammer, T., Estrin, D., Hansen, M., Harmon, T., Kohler, E., and Srivastava, M. 2006b. The final frontier: Embedding networked sensors in the soil. Tech. rep. 68, CENS.Google Scholar
- SAS. The SAS forecasting software. http://www.sas.com/technologies/analytics/forecasting/index.html.Google Scholar
- SensorScope. 2006. The SensorScope Lausanne urban canopy experiment (LUCE) project. Dataset available at http://sensorscope.epfl.ch/index.php/LUCE.Google Scholar
- Szewczyk, R., Polastre, J., Mainwaring, A., and Culler, D. 2004. Lessons from a sensor network expedition. In Proceedings of the 1st European Workshop on Sensor Networks (EWSN).Google Scholar
- Tolle, G., Polastre, J., Szewczyk, R., Culler, D., Turner, N., Tu, K., Burgess, S., Dawson, T., Buonadonna, P., Gay, D., and Hong, W. 2005. A macroscope in the redwoods. In Proceedings of the 2nd International Conference on Embedded Networked Sensor Systems (SenSys). ACM Press, New York, 51--63. Google ScholarDigital Library
- Tulone, D. and Madden, S. 2006. PAQ: Time series forecasting for approximate query answering in sensor networks. In Proceedings of the European Conference on Wireless Sensor Networks (EWSN). Google ScholarDigital Library
- Werner-Allen, G., Lorincz, K., Johnson, J., Lees, J., and Welsh, M. 2006. Fidelity and yield in a volcano monitoring sensor network. In Proceedings of the 7th USENIX Symposium on Operating Systems Design and Implementation (OSDI). Google ScholarDigital Library
- Zahedi, S., Szczodrak, M., Ji, P., Mylaraswamy, D., Srivastava, M. B., and Young, R. 2008. Tiered architecture for on-line detection, isolation and repair of faults in wireless sensor networks. In Proceedings of the MILCOM.Google Scholar
Index Terms
- Sensor faults: Detection methods and prevalence in real-world datasets
Recommendations
A distributed adaptive scheme for detecting faults in wireless sensor networks
This paper presents a distributed adaptive scheme for detecting faults in wireless sensor networks. Each sensor nodes makes a local decision based on the comparisons of its own readings with those of neighbors, along with the dissemination of the ...
A Survey of Fault Management in Wireless Sensor Networks
Wireless sensor networks are resource-constrained self-organizing systems that are often deployed in inaccessible and inhospitable environments in order to collect data about some outside world phenomenon. For most sensor network applications, point-to-...
Checkpoint Faults are not Sufficient Target Faults for Test Generation
Stuck-at faults on primary inputs and fan-out branches are commonly used as target faults in test generation algorithms for combinational circuits. This correspondence shows that these faults may not constitute an adequate set of target faults. A ...
Comments