Classification of reactor facility operational state using SPRT methods with radiation sensor networks

Camila Ramirez; Nageswara S.V. Rao

doi:10.1007/978-3-319-90509-9_5

Classification of reactor facility operational state using SPRT methods with radiation sensor networks

Camila Ramirez, Nageswara S.V. Rao

Autonomous and Complex Systems

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

We consider the problem of inferring the operational state of a reactor facility by using measurements from a radiation sensor network, which is deployed around the facility’s ventilation stack. The radiation emissions from the stack decay with distance, and the corresponding measurements are inherently random with parameters determined by radiation intensity levels at the sensor locations. We fuse measurements from network sensors to estimate the intensity at the stack, and use this estimate in a one-sided Sequential Probability Ratio Test (SPRT) to infer the on/off state of the reactor facility. We demonstrate the superior performance of this method over conventional majority vote fusers and individual sensors using (i) test measurements from a network of NaI sensors, and (ii) emulated measurements using radioactive effluents collected at a reactor facility stack. We analytically quantify the performance improvements of individual sensors and their networks with adaptive thresholds over those with fixed ones, by using the packing number of the radiation intensity space.

Original language	English
Title of host publication	Multisensor Fusion and Integration in the Wake of Big Data, Deep Learning and Cyber Physical System - An Edition of the Selected Papers from the 2017 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems MFI 2017
Editors	Hanseok Ko, Sukhan Lee, Songhwai Oh
Publisher	Springer Verlag
Pages	76-97
Number of pages	22
ISBN (Print)	9783319905082
DOIs	https://doi.org/10.1007/978-3-319-90509-9_5
State	Published - 2018
Event	13th IEEE International Conference on Multisensor Integration and Fusion, IEEE MFI 2017 - Daegu, Korea, Republic of Duration: Nov 16 2017 → Nov 22 2017

Publication series

Name	Lecture Notes in Electrical Engineering
Volume	501
ISSN (Print)	1876-1100
ISSN (Electronic)	1876-1119

Conference

Conference	13th IEEE International Conference on Multisensor Integration and Fusion, IEEE MFI 2017
Country/Territory	Korea, Republic of
City	Daegu
Period	11/16/17 → 11/22/17

Keywords

Localization
Radiation sensor network
Reactor facility
Sequential probability ratio test
State detection

Access to Document

10.1007/978-3-319-90509-9_5

Cite this

Ramirez, C., & Rao, N. S. V. (2018). Classification of reactor facility operational state using SPRT methods with radiation sensor networks. In H. Ko, S. Lee, & S. Oh (Eds.), Multisensor Fusion and Integration in the Wake of Big Data, Deep Learning and Cyber Physical System - An Edition of the Selected Papers from the 2017 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems MFI 2017 (pp. 76-97). (Lecture Notes in Electrical Engineering; Vol. 501). Springer Verlag. https://doi.org/10.1007/978-3-319-90509-9_5

Ramirez, Camila ; Rao, Nageswara S.V. / Classification of reactor facility operational state using SPRT methods with radiation sensor networks. Multisensor Fusion and Integration in the Wake of Big Data, Deep Learning and Cyber Physical System - An Edition of the Selected Papers from the 2017 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems MFI 2017. editor / Hanseok Ko ; Sukhan Lee ; Songhwai Oh. Springer Verlag, 2018. pp. 76-97 (Lecture Notes in Electrical Engineering).

@inproceedings{5eb97bd619e44d94ba4b03170bde2c1c,

title = "Classification of reactor facility operational state using SPRT methods with radiation sensor networks",

abstract = "We consider the problem of inferring the operational state of a reactor facility by using measurements from a radiation sensor network, which is deployed around the facility{\textquoteright}s ventilation stack. The radiation emissions from the stack decay with distance, and the corresponding measurements are inherently random with parameters determined by radiation intensity levels at the sensor locations. We fuse measurements from network sensors to estimate the intensity at the stack, and use this estimate in a one-sided Sequential Probability Ratio Test (SPRT) to infer the on/off state of the reactor facility. We demonstrate the superior performance of this method over conventional majority vote fusers and individual sensors using (i) test measurements from a network of NaI sensors, and (ii) emulated measurements using radioactive effluents collected at a reactor facility stack. We analytically quantify the performance improvements of individual sensors and their networks with adaptive thresholds over those with fixed ones, by using the packing number of the radiation intensity space.",

keywords = "Localization, Radiation sensor network, Reactor facility, Sequential probability ratio test, State detection",

author = "Camila Ramirez and Rao, {Nageswara S.V.}",

note = "Publisher Copyright: {\textcopyright} 2018, Springer International Publishing AG, part of Springer Nature.; 13th IEEE International Conference on Multisensor Integration and Fusion, IEEE MFI 2017 ; Conference date: 16-11-2017 Through 22-11-2017",

year = "2018",

doi = "10.1007/978-3-319-90509-9_5",

language = "English",

isbn = "9783319905082",

series = "Lecture Notes in Electrical Engineering",

publisher = "Springer Verlag",

pages = "76--97",

editor = "Hanseok Ko and Sukhan Lee and Songhwai Oh",

booktitle = "Multisensor Fusion and Integration in the Wake of Big Data, Deep Learning and Cyber Physical System - An Edition of the Selected Papers from the 2017 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems MFI 2017",

}

Ramirez, C & Rao, NSV 2018, Classification of reactor facility operational state using SPRT methods with radiation sensor networks. in H Ko, S Lee & S Oh (eds), Multisensor Fusion and Integration in the Wake of Big Data, Deep Learning and Cyber Physical System - An Edition of the Selected Papers from the 2017 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems MFI 2017. Lecture Notes in Electrical Engineering, vol. 501, Springer Verlag, pp. 76-97, 13th IEEE International Conference on Multisensor Integration and Fusion, IEEE MFI 2017, Daegu, Korea, Republic of, 11/16/17. https://doi.org/10.1007/978-3-319-90509-9_5

Classification of reactor facility operational state using SPRT methods with radiation sensor networks. / Ramirez, Camila; Rao, Nageswara S.V.
Multisensor Fusion and Integration in the Wake of Big Data, Deep Learning and Cyber Physical System - An Edition of the Selected Papers from the 2017 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems MFI 2017. ed. / Hanseok Ko; Sukhan Lee; Songhwai Oh. Springer Verlag, 2018. p. 76-97 (Lecture Notes in Electrical Engineering; Vol. 501).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Classification of reactor facility operational state using SPRT methods with radiation sensor networks

AU - Ramirez, Camila

AU - Rao, Nageswara S.V.

PY - 2018

Y1 - 2018

N2 - We consider the problem of inferring the operational state of a reactor facility by using measurements from a radiation sensor network, which is deployed around the facility’s ventilation stack. The radiation emissions from the stack decay with distance, and the corresponding measurements are inherently random with parameters determined by radiation intensity levels at the sensor locations. We fuse measurements from network sensors to estimate the intensity at the stack, and use this estimate in a one-sided Sequential Probability Ratio Test (SPRT) to infer the on/off state of the reactor facility. We demonstrate the superior performance of this method over conventional majority vote fusers and individual sensors using (i) test measurements from a network of NaI sensors, and (ii) emulated measurements using radioactive effluents collected at a reactor facility stack. We analytically quantify the performance improvements of individual sensors and their networks with adaptive thresholds over those with fixed ones, by using the packing number of the radiation intensity space.

AB - We consider the problem of inferring the operational state of a reactor facility by using measurements from a radiation sensor network, which is deployed around the facility’s ventilation stack. The radiation emissions from the stack decay with distance, and the corresponding measurements are inherently random with parameters determined by radiation intensity levels at the sensor locations. We fuse measurements from network sensors to estimate the intensity at the stack, and use this estimate in a one-sided Sequential Probability Ratio Test (SPRT) to infer the on/off state of the reactor facility. We demonstrate the superior performance of this method over conventional majority vote fusers and individual sensors using (i) test measurements from a network of NaI sensors, and (ii) emulated measurements using radioactive effluents collected at a reactor facility stack. We analytically quantify the performance improvements of individual sensors and their networks with adaptive thresholds over those with fixed ones, by using the packing number of the radiation intensity space.

KW - Localization

KW - Radiation sensor network

KW - Reactor facility

KW - Sequential probability ratio test

KW - State detection

UR - http://www.scopus.com/inward/record.url?scp=85049977786&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-90509-9_5

DO - 10.1007/978-3-319-90509-9_5

M3 - Conference contribution

AN - SCOPUS:85049977786

SN - 9783319905082

T3 - Lecture Notes in Electrical Engineering

SP - 76

EP - 97

BT - Multisensor Fusion and Integration in the Wake of Big Data, Deep Learning and Cyber Physical System - An Edition of the Selected Papers from the 2017 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems MFI 2017

A2 - Ko, Hanseok

A2 - Lee, Sukhan

A2 - Oh, Songhwai

PB - Springer Verlag

T2 - 13th IEEE International Conference on Multisensor Integration and Fusion, IEEE MFI 2017

Y2 - 16 November 2017 through 22 November 2017

ER -

Ramirez C, Rao NSV. Classification of reactor facility operational state using SPRT methods with radiation sensor networks. In Ko H, Lee S, Oh S, editors, Multisensor Fusion and Integration in the Wake of Big Data, Deep Learning and Cyber Physical System - An Edition of the Selected Papers from the 2017 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems MFI 2017. Springer Verlag. 2018. p. 76-97. (Lecture Notes in Electrical Engineering). doi: 10.1007/978-3-319-90509-9_5

Classification of reactor facility operational state using SPRT methods with radiation sensor networks

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this