Achievable distortion/rate tradeoff in a decentralized Gaussian parameter estimation problem

Anna Scaglione, Mehmet E. Yildiz, Tuncer C. Aysal

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we study the decentralized version of the classical rate constrained Gaussian parameter estimation problem referred to as the Central Estimation Officer (CEO) problem which we refer to as the Decentralized Estimation Officers (DEO) problem. Like in the CEO case, we consider a group of N sensors observing an independently corrupted version of an infinite i.i.d. sequence of samples from a Gaussian source, in additive Gaussian noise. Unlike the CEO case, the sensors in our study are also the estimation officers. They are uniformly deployed in a circular pattern of radius r and communicate over RF links with limited energy. Their task is to reconstruct the quantity of interest (the samples of the source), without a central fusion node, better than what they are capable of with their local observations. We find achievable scaling laws by structuring our communication protocol as an instance of the so called average consensus algorithm, a gossiping protocol used for averaging original sensor measurements via near neighbors communications. We derive how the Mean Squared Error (MSE) of the sensors' estimation scales with the network size, per node power and ring radius r. Moreover, we compare our results with scaling laws previously derived for the centralized case, i.e, the CEO problem in a comparable scenario.

Original languageEnglish (US)
Title of host publicationInternational Zurich Seminar on Digital Communications
Pages102-105
Number of pages4
DOIs
StatePublished - 2008
Externally publishedYes
Event2008 International Zurich Seminar on Communications, IZS - Zurich, Switzerland
Duration: Mar 12 2008Mar 14 2008

Other

Other2008 International Zurich Seminar on Communications, IZS
CountrySwitzerland
CityZurich
Period3/12/083/14/08

Fingerprint

Parameter estimation
Scaling laws
Sensors
Network protocols
Fusion reactions
Communication

Keywords

  • Average consensus
  • CEO problem
  • Parameter estimation
  • Sensor networks

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Scaglione, A., Yildiz, M. E., & Aysal, T. C. (2008). Achievable distortion/rate tradeoff in a decentralized Gaussian parameter estimation problem. In International Zurich Seminar on Digital Communications (pp. 102-105). [4497286] https://doi.org/10.1109/IZS.2008.4497286

Achievable distortion/rate tradeoff in a decentralized Gaussian parameter estimation problem. / Scaglione, Anna; Yildiz, Mehmet E.; Aysal, Tuncer C.

International Zurich Seminar on Digital Communications. 2008. p. 102-105 4497286.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Scaglione, A, Yildiz, ME & Aysal, TC 2008, Achievable distortion/rate tradeoff in a decentralized Gaussian parameter estimation problem. in International Zurich Seminar on Digital Communications., 4497286, pp. 102-105, 2008 International Zurich Seminar on Communications, IZS, Zurich, Switzerland, 3/12/08. https://doi.org/10.1109/IZS.2008.4497286
Scaglione A, Yildiz ME, Aysal TC. Achievable distortion/rate tradeoff in a decentralized Gaussian parameter estimation problem. In International Zurich Seminar on Digital Communications. 2008. p. 102-105. 4497286 https://doi.org/10.1109/IZS.2008.4497286
Scaglione, Anna ; Yildiz, Mehmet E. ; Aysal, Tuncer C. / Achievable distortion/rate tradeoff in a decentralized Gaussian parameter estimation problem. International Zurich Seminar on Digital Communications. 2008. pp. 102-105
@inproceedings{e7f54ab4077d4e0f90984b5b4ce85404,
title = "Achievable distortion/rate tradeoff in a decentralized Gaussian parameter estimation problem",
abstract = "In this paper, we study the decentralized version of the classical rate constrained Gaussian parameter estimation problem referred to as the Central Estimation Officer (CEO) problem which we refer to as the Decentralized Estimation Officers (DEO) problem. Like in the CEO case, we consider a group of N sensors observing an independently corrupted version of an infinite i.i.d. sequence of samples from a Gaussian source, in additive Gaussian noise. Unlike the CEO case, the sensors in our study are also the estimation officers. They are uniformly deployed in a circular pattern of radius r and communicate over RF links with limited energy. Their task is to reconstruct the quantity of interest (the samples of the source), without a central fusion node, better than what they are capable of with their local observations. We find achievable scaling laws by structuring our communication protocol as an instance of the so called average consensus algorithm, a gossiping protocol used for averaging original sensor measurements via near neighbors communications. We derive how the Mean Squared Error (MSE) of the sensors' estimation scales with the network size, per node power and ring radius r. Moreover, we compare our results with scaling laws previously derived for the centralized case, i.e, the CEO problem in a comparable scenario.",
keywords = "Average consensus, CEO problem, Parameter estimation, Sensor networks",
author = "Anna Scaglione and Yildiz, {Mehmet E.} and Aysal, {Tuncer C.}",
year = "2008",
doi = "10.1109/IZS.2008.4497286",
language = "English (US)",
isbn = "9781424416820",
pages = "102--105",
booktitle = "International Zurich Seminar on Digital Communications",

}

TY - GEN

T1 - Achievable distortion/rate tradeoff in a decentralized Gaussian parameter estimation problem

AU - Scaglione, Anna

AU - Yildiz, Mehmet E.

AU - Aysal, Tuncer C.

PY - 2008

Y1 - 2008

N2 - In this paper, we study the decentralized version of the classical rate constrained Gaussian parameter estimation problem referred to as the Central Estimation Officer (CEO) problem which we refer to as the Decentralized Estimation Officers (DEO) problem. Like in the CEO case, we consider a group of N sensors observing an independently corrupted version of an infinite i.i.d. sequence of samples from a Gaussian source, in additive Gaussian noise. Unlike the CEO case, the sensors in our study are also the estimation officers. They are uniformly deployed in a circular pattern of radius r and communicate over RF links with limited energy. Their task is to reconstruct the quantity of interest (the samples of the source), without a central fusion node, better than what they are capable of with their local observations. We find achievable scaling laws by structuring our communication protocol as an instance of the so called average consensus algorithm, a gossiping protocol used for averaging original sensor measurements via near neighbors communications. We derive how the Mean Squared Error (MSE) of the sensors' estimation scales with the network size, per node power and ring radius r. Moreover, we compare our results with scaling laws previously derived for the centralized case, i.e, the CEO problem in a comparable scenario.

AB - In this paper, we study the decentralized version of the classical rate constrained Gaussian parameter estimation problem referred to as the Central Estimation Officer (CEO) problem which we refer to as the Decentralized Estimation Officers (DEO) problem. Like in the CEO case, we consider a group of N sensors observing an independently corrupted version of an infinite i.i.d. sequence of samples from a Gaussian source, in additive Gaussian noise. Unlike the CEO case, the sensors in our study are also the estimation officers. They are uniformly deployed in a circular pattern of radius r and communicate over RF links with limited energy. Their task is to reconstruct the quantity of interest (the samples of the source), without a central fusion node, better than what they are capable of with their local observations. We find achievable scaling laws by structuring our communication protocol as an instance of the so called average consensus algorithm, a gossiping protocol used for averaging original sensor measurements via near neighbors communications. We derive how the Mean Squared Error (MSE) of the sensors' estimation scales with the network size, per node power and ring radius r. Moreover, we compare our results with scaling laws previously derived for the centralized case, i.e, the CEO problem in a comparable scenario.

KW - Average consensus

KW - CEO problem

KW - Parameter estimation

KW - Sensor networks

UR - http://www.scopus.com/inward/record.url?scp=51349094385&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=51349094385&partnerID=8YFLogxK

U2 - 10.1109/IZS.2008.4497286

DO - 10.1109/IZS.2008.4497286

M3 - Conference contribution

SN - 9781424416820

SP - 102

EP - 105

BT - International Zurich Seminar on Digital Communications

ER -