The embedded triangles algorithm for distributed estimation in sensor networks

V. Delouille; R. Neelamani; V. Chandrasekaran; R. G. Baraniuk

doi:10.1109/SSP.2003.1289422

The embedded triangles algorithm for distributed estimation in sensor networks

V. Delouille, R. Neelamani, V. Chandrasekaran, R. G. Baraniuk

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

8 Scopus citations

Abstract

We propose a new iterative distributed estimation algorithm for Gaussian hidden Markov graphical models with loops. We decompose a loopy graph into a number of linked embedded triangles and then apply a parallel block-Jacobi iteration comprising local linear minimum mean-square-error estimation on each triangle (involving a simple 3x3 matrix inverse computation) followed by an information exchange between neighboring nodes and triangles. A simulation study demonstrates that the algorithm converges extremely rapidly, outperforming a number of existing algorithms. Embedded triangles are simple, local, scalable, fault-tolerant, and energy-efficient, and thus ideally suited for wireless sensor networks.

Original language	English (US)
Title of host publication	IEEE Workshop on Statistical Signal Processing Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	371-374
Number of pages	4
Volume	2003-January
ISBN (Print)	0780379977
DOIs	https://doi.org/10.1109/SSP.2003.1289422
State	Published - 2003
Event	IEEE Workshop on Statistical Signal Processing, SSP 2003 - St. Louis, United States Duration: Sep 28 2003 → Oct 1 2003

Other

Other	IEEE Workshop on Statistical Signal Processing, SSP 2003
Country	United States
City	St. Louis
Period	9/28/03 → 10/1/03

Keywords

Computational modeling
Concurrent computing
Embedded computing
Energy efficiency
Fault tolerance
Graphical models
Hidden Markov models
Iterative algorithms
Matrix decomposition
Wireless sensor networks

ASJC Scopus subject areas

Electrical and Electronic Engineering
Applied Mathematics
Signal Processing
Computer Science Applications

Access to Document

10.1109/SSP.2003.1289422

Fingerprint Dive into the research topics of 'The embedded triangles algorithm for distributed estimation in sensor networks'. Together they form a unique fingerprint.

Cite this

The embedded triangles algorithm for distributed estimation in sensor networks. / Delouille, V.; Neelamani, R.; Chandrasekaran, V.; Baraniuk, R. G.

IEEE Workshop on Statistical Signal Processing Proceedings. Vol. 2003-January Institute of Electrical and Electronics Engineers Inc., 2003. p. 371-374 1289422.

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

Delouille, V, Neelamani, R, Chandrasekaran, V & Baraniuk, RG 2003, The embedded triangles algorithm for distributed estimation in sensor networks. in IEEE Workshop on Statistical Signal Processing Proceedings. vol. 2003-January, 1289422, Institute of Electrical and Electronics Engineers Inc., pp. 371-374, IEEE Workshop on Statistical Signal Processing, SSP 2003, St. Louis, United States, 9/28/03. https://doi.org/10.1109/SSP.2003.1289422

@inproceedings{fb5f13666ac84774a544f67b17b59fdb,

title = "The embedded triangles algorithm for distributed estimation in sensor networks",

abstract = "We propose a new iterative distributed estimation algorithm for Gaussian hidden Markov graphical models with loops. We decompose a loopy graph into a number of linked embedded triangles and then apply a parallel block-Jacobi iteration comprising local linear minimum mean-square-error estimation on each triangle (involving a simple 3x3 matrix inverse computation) followed by an information exchange between neighboring nodes and triangles. A simulation study demonstrates that the algorithm converges extremely rapidly, outperforming a number of existing algorithms. Embedded triangles are simple, local, scalable, fault-tolerant, and energy-efficient, and thus ideally suited for wireless sensor networks.",

keywords = "Computational modeling, Concurrent computing, Embedded computing, Energy efficiency, Fault tolerance, Graphical models, Hidden Markov models, Iterative algorithms, Matrix decomposition, Wireless sensor networks",

author = "V. Delouille and R. Neelamani and V. Chandrasekaran and Baraniuk, {R. G.}",

year = "2003",

doi = "10.1109/SSP.2003.1289422",

language = "English (US)",

isbn = "0780379977",

volume = "2003-January",

pages = "371--374",

booktitle = "IEEE Workshop on Statistical Signal Processing Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

address = "United States",

note = "IEEE Workshop on Statistical Signal Processing, SSP 2003 ; Conference date: 28-09-2003 Through 01-10-2003",

}

TY - GEN

T1 - The embedded triangles algorithm for distributed estimation in sensor networks

AU - Delouille, V.

AU - Neelamani, R.

AU - Chandrasekaran, V.

AU - Baraniuk, R. G.

PY - 2003

Y1 - 2003

N2 - We propose a new iterative distributed estimation algorithm for Gaussian hidden Markov graphical models with loops. We decompose a loopy graph into a number of linked embedded triangles and then apply a parallel block-Jacobi iteration comprising local linear minimum mean-square-error estimation on each triangle (involving a simple 3x3 matrix inverse computation) followed by an information exchange between neighboring nodes and triangles. A simulation study demonstrates that the algorithm converges extremely rapidly, outperforming a number of existing algorithms. Embedded triangles are simple, local, scalable, fault-tolerant, and energy-efficient, and thus ideally suited for wireless sensor networks.

AB - We propose a new iterative distributed estimation algorithm for Gaussian hidden Markov graphical models with loops. We decompose a loopy graph into a number of linked embedded triangles and then apply a parallel block-Jacobi iteration comprising local linear minimum mean-square-error estimation on each triangle (involving a simple 3x3 matrix inverse computation) followed by an information exchange between neighboring nodes and triangles. A simulation study demonstrates that the algorithm converges extremely rapidly, outperforming a number of existing algorithms. Embedded triangles are simple, local, scalable, fault-tolerant, and energy-efficient, and thus ideally suited for wireless sensor networks.

KW - Computational modeling

KW - Concurrent computing

KW - Embedded computing

KW - Energy efficiency

KW - Fault tolerance

KW - Graphical models

KW - Hidden Markov models

KW - Iterative algorithms

KW - Matrix decomposition

KW - Wireless sensor networks

UR - http://www.scopus.com/inward/record.url?scp=84948691099&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84948691099&partnerID=8YFLogxK

U2 - 10.1109/SSP.2003.1289422

DO - 10.1109/SSP.2003.1289422

M3 - Conference contribution

AN - SCOPUS:84948691099

SN - 0780379977

VL - 2003-January

SP - 371

EP - 374

BT - IEEE Workshop on Statistical Signal Processing Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - IEEE Workshop on Statistical Signal Processing, SSP 2003

Y2 - 28 September 2003 through 1 October 2003

ER -