Quantized variational filtering for bayesian inference in wireless sensor networks

Majdi Mansouri, Hichem Snoussi, Jing Teng, Ouachani Ilham, Cédric Richard

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

The primary focus of the chapter is to study the Bayesian inference problem in distributed WSNs with particular emphasis on the trade-off between estimation precision and energy-awareness. We propose a variational approach to approximate the particle distribution to a single Gaussian distribution, while respecting the communication constraints of WSNs. The efficiency of the variational approximation relies on the fact that the online update and the compression of the filtering distribution are simultaneously performed. In addition, the variational approach has the nice property to be model-free, ensuring robustness of signal processing. We analyze the Bayesian inference issue for several specific but representative WSN applications to elaborate the quantized variational filtering method, which can be applicable to a wider class of problems.

Original languageEnglish
Title of host publicationVisual Information Processing in Wireless Sensor Networks: Technology, Trends and Applications
PublisherIGI Global
Pages224-249
Number of pages26
ISBN (Print)9781613501535
DOIs
Publication statusPublished - 1800
Externally publishedYes

Fingerprint

Wireless sensor networks
Gaussian distribution
Signal processing
Communication

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Mansouri, M., Snoussi, H., Teng, J., Ilham, O., & Richard, C. (1800). Quantized variational filtering for bayesian inference in wireless sensor networks. In Visual Information Processing in Wireless Sensor Networks: Technology, Trends and Applications (pp. 224-249). IGI Global. https://doi.org/10.4018/978-1-61350-153-5.ch011

Quantized variational filtering for bayesian inference in wireless sensor networks. / Mansouri, Majdi; Snoussi, Hichem; Teng, Jing; Ilham, Ouachani; Richard, Cédric.

Visual Information Processing in Wireless Sensor Networks: Technology, Trends and Applications. IGI Global, 1800. p. 224-249.

Research output: Chapter in Book/Report/Conference proceedingChapter

Mansouri, M, Snoussi, H, Teng, J, Ilham, O & Richard, C 1800, Quantized variational filtering for bayesian inference in wireless sensor networks. in Visual Information Processing in Wireless Sensor Networks: Technology, Trends and Applications. IGI Global, pp. 224-249. https://doi.org/10.4018/978-1-61350-153-5.ch011
Mansouri M, Snoussi H, Teng J, Ilham O, Richard C. Quantized variational filtering for bayesian inference in wireless sensor networks. In Visual Information Processing in Wireless Sensor Networks: Technology, Trends and Applications. IGI Global. 1800. p. 224-249 https://doi.org/10.4018/978-1-61350-153-5.ch011
Mansouri, Majdi ; Snoussi, Hichem ; Teng, Jing ; Ilham, Ouachani ; Richard, Cédric. / Quantized variational filtering for bayesian inference in wireless sensor networks. Visual Information Processing in Wireless Sensor Networks: Technology, Trends and Applications. IGI Global, 1800. pp. 224-249
@inbook{fcd2b790583b4c73a402c35957302a16,
title = "Quantized variational filtering for bayesian inference in wireless sensor networks",
abstract = "The primary focus of the chapter is to study the Bayesian inference problem in distributed WSNs with particular emphasis on the trade-off between estimation precision and energy-awareness. We propose a variational approach to approximate the particle distribution to a single Gaussian distribution, while respecting the communication constraints of WSNs. The efficiency of the variational approximation relies on the fact that the online update and the compression of the filtering distribution are simultaneously performed. In addition, the variational approach has the nice property to be model-free, ensuring robustness of signal processing. We analyze the Bayesian inference issue for several specific but representative WSN applications to elaborate the quantized variational filtering method, which can be applicable to a wider class of problems.",
author = "Majdi Mansouri and Hichem Snoussi and Jing Teng and Ouachani Ilham and C{\'e}dric Richard",
year = "1800",
doi = "10.4018/978-1-61350-153-5.ch011",
language = "English",
isbn = "9781613501535",
pages = "224--249",
booktitle = "Visual Information Processing in Wireless Sensor Networks: Technology, Trends and Applications",
publisher = "IGI Global",

}

TY - CHAP

T1 - Quantized variational filtering for bayesian inference in wireless sensor networks

AU - Mansouri, Majdi

AU - Snoussi, Hichem

AU - Teng, Jing

AU - Ilham, Ouachani

AU - Richard, Cédric

PY - 1800

Y1 - 1800

N2 - The primary focus of the chapter is to study the Bayesian inference problem in distributed WSNs with particular emphasis on the trade-off between estimation precision and energy-awareness. We propose a variational approach to approximate the particle distribution to a single Gaussian distribution, while respecting the communication constraints of WSNs. The efficiency of the variational approximation relies on the fact that the online update and the compression of the filtering distribution are simultaneously performed. In addition, the variational approach has the nice property to be model-free, ensuring robustness of signal processing. We analyze the Bayesian inference issue for several specific but representative WSN applications to elaborate the quantized variational filtering method, which can be applicable to a wider class of problems.

AB - The primary focus of the chapter is to study the Bayesian inference problem in distributed WSNs with particular emphasis on the trade-off between estimation precision and energy-awareness. We propose a variational approach to approximate the particle distribution to a single Gaussian distribution, while respecting the communication constraints of WSNs. The efficiency of the variational approximation relies on the fact that the online update and the compression of the filtering distribution are simultaneously performed. In addition, the variational approach has the nice property to be model-free, ensuring robustness of signal processing. We analyze the Bayesian inference issue for several specific but representative WSN applications to elaborate the quantized variational filtering method, which can be applicable to a wider class of problems.

UR - http://www.scopus.com/inward/record.url?scp=84898505745&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84898505745&partnerID=8YFLogxK

U2 - 10.4018/978-1-61350-153-5.ch011

DO - 10.4018/978-1-61350-153-5.ch011

M3 - Chapter

SN - 9781613501535

SP - 224

EP - 249

BT - Visual Information Processing in Wireless Sensor Networks: Technology, Trends and Applications

PB - IGI Global

ER -