A Networked Evidence Theory Framework for Critical Infrastructure Modeling

Abstract : This paper describes a distributed approach for data fusion and information sharing based on evidence theory and the transferable belief model. Evidence theory aggregates data generated from different sources in order to better assess an ongoing situation and to aid in the response and decision making processes. In the domain of critical infrastructure protection, researchers are forced to develop distributed approaches for modeling and control with a minimal exchange of data due to the existence of multiple stakeholders and interconnections between infrastructure components. Evidence theory permits the modeling of uncertainty in data fusion, but it is typically applied in a centralized manner. This paper proposes a decentralized extension of the transferable belief model that facilitates the application of evidence theory to data fusion in critical infrastructure applications. A case study is provided to demonstrate the convergence of results similar to the centralized approach, and to show the utility of fusing data in a distributed manner for interdependent critical infrastructure systems.
Document type :
Conference papers
Complete list of metadatas

Cited literature [12 references]  Display  Hide  Download

https://hal.inria.fr/hal-01483814
Contributor : Hal Ifip <>
Submitted on : Monday, March 6, 2017 - 3:01:51 PM
Last modification on : Monday, March 6, 2017 - 3:05:07 PM
Long-term archiving on : Wednesday, June 7, 2017 - 2:33:58 PM

File

978-3-642-35764-0_15_Chapter.p...
Files produced by the author(s)

Licence


Distributed under a Creative Commons Attribution 4.0 International License

Identifiers

Citation

Chiara Foglietta, Andrea Gasparri, Stefano Panzieri. A Networked Evidence Theory Framework for Critical Infrastructure Modeling. 6th International Conference on Critical Infrastructure Protection (ICCIP), Mar 2012, Washington, DC, United States. pp.205-215, ⟨10.1007/978-3-642-35764-0_15⟩. ⟨hal-01483814⟩

Share

Metrics

Record views

101

Files downloads

88