An Internet of Lying Things: Probabilistic fault detection of nonverifiable sensors


We are experiencing an explosion in the amount of sensors measuring our activities and the world around us. These sensors are spread throughout the built environment and can help us perform state estimation and control of related systems, but they are often built and/or maintained by third parties or system users. As a result, by outsourcing system measurement to third parties, the controller must accept their measurements without being able to directly verify the sensors’ correct operation. Instead, detection and rejection of measurements from faulty sensors must be done with the raw data only. Towards this goal, we present a probabilistic formulation of state estimation with a model for possibly faulty behavior of sensors. We also take into consideration the possibility that the control designer may not know the characteristics of faulty measurements, and discuss how they may be alternatively detected by how much they differ from “expected” measurements. We detail implementation of the probabilistic formalism in a particle filtering application. Finally, we present results that use these methods, where the state of road traffic on a freeway is estimated via a particle filter by fusing third-party global navigational satellite system readings, while rejecting faulty measurements. The results demonstrate that faulty third-party measurements may be detected and removed without explicit models of a fault’s characteristics.

Extracted Key Phrases

4 Figures and Tables

Cite this paper

@article{Wright2016AnIO, title={An Internet of Lying Things: Probabilistic fault detection of nonverifiable sensors}, author={Matthew A. Wright and Roberto Horowitz}, journal={CoRR}, year={2016}, volume={abs/1609.06795} }