Distributed estimation using reduced-dimensionality sensor observations

Ioannis D. Schizas; Georgios B. Giannakis; Zhi Quan Luo

doi:10.1109/TSP.2007.895987

Distributed estimation using reduced-dimensionality sensor observations

Ioannis D. Schizas, Georgios B. Giannakis, Zhi Quan Luo

Electrical and Computer Engineering

Research output: Contribution to journal › Article › peer-review

178 Scopus citations

Abstract

We derive linear estimators of stationary random signals based on reduced-dimensionality observations collected at distributed sensors and communicated to a fusion center over wireless links. Dimensionality reduction compresses sensor data to meet low-power and bandwidth constraints, while linearity in compression and estimation are well motivated by the limited computing capabilities wireless sensor networks are envisioned to operate with, and by the desire to estimate random signals from observations with unknown probability density functions. In the absence of fading and fusion center noise (ideal links), we cast this intertwined compression-estimation problem in a canonical correlation analysis framework and derive closed-form mean-square error (MSE) optimal estimators along with coordinate descent suboptimal alternatives that guarantee convergence at least to a stationary point. Likewise, we develop estimators based on reduced-dimensionality sensor observations in the presence of fading and additive noise at the fusion center (nonideal links). Performance analysis and corroborating simulations demonstrate the merits of the novel distributed estimators relative to existing alternatives.

Original language	English (US)
Pages (from-to)	4284-4299
Number of pages	16
Journal	IEEE Transactions on Signal Processing
Volume	55
Issue number	8
DOIs	https://doi.org/10.1109/TSP.2007.895987
State	Published - Aug 2007

Bibliographical note

Funding Information:
Manuscript received November 6, 2005; revised August 10, 2006. The associate editor coordinating the review of this manuscript and approving it for publication was Prof. Javier Garcia-Frias. Prepared through collaborative participation in the Communications and Networks Consortium sponsored by the U. S. Army Research Laboratory under the Collaborative Technology Alliance Program, Cooperative Agreement DAAD19-01-2-0011. The work of Z.-Q. Luo is supported by the U.S. DoD Army, grant number W911NF-05-1-0567. The U.S. Government is authorized to reproduce and distribute reprints for Government purposes notwithstanding any copyright notation thereon. Parts of the paper were presented at the Thirty-Ninth Asilomar Conference, Pacific Grove, CA, Oct. 30–November 2, 2005, and at the 2006 International Conference on Acoustics, Speech and Signal Processing (ICASSP), Toulouse, France, May 14–19, 2006.

Keywords

Canonical correlation analysis (CCA)
Distributed compression
Distributed estimation
Nonlinear optimization
Wireless sensor networks (WSNs)

Access

10.1109/TSP.2007.895987

OpenUrl availability

Full text

Cite this

@article{b5fe571a47544ce9bd31bed4f563c23d,

title = "Distributed estimation using reduced-dimensionality sensor observations",

abstract = "We derive linear estimators of stationary random signals based on reduced-dimensionality observations collected at distributed sensors and communicated to a fusion center over wireless links. Dimensionality reduction compresses sensor data to meet low-power and bandwidth constraints, while linearity in compression and estimation are well motivated by the limited computing capabilities wireless sensor networks are envisioned to operate with, and by the desire to estimate random signals from observations with unknown probability density functions. In the absence of fading and fusion center noise (ideal links), we cast this intertwined compression-estimation problem in a canonical correlation analysis framework and derive closed-form mean-square error (MSE) optimal estimators along with coordinate descent suboptimal alternatives that guarantee convergence at least to a stationary point. Likewise, we develop estimators based on reduced-dimensionality sensor observations in the presence of fading and additive noise at the fusion center (nonideal links). Performance analysis and corroborating simulations demonstrate the merits of the novel distributed estimators relative to existing alternatives.",

keywords = "Canonical correlation analysis (CCA), Distributed compression, Distributed estimation, Nonlinear optimization, Wireless sensor networks (WSNs)",

author = "Schizas, {Ioannis D.} and Giannakis, {Georgios B.} and Luo, {Zhi Quan}",

note = "Funding Information: Manuscript received November 6, 2005; revised August 10, 2006. The associate editor coordinating the review of this manuscript and approving it for publication was Prof. Javier Garcia-Frias. Prepared through collaborative participation in the Communications and Networks Consortium sponsored by the U. S. Army Research Laboratory under the Collaborative Technology Alliance Program, Cooperative Agreement DAAD19-01-2-0011. The work of Z.-Q. Luo is supported by the U.S. DoD Army, grant number W911NF-05-1-0567. The U.S. Government is authorized to reproduce and distribute reprints for Government purposes notwithstanding any copyright notation thereon. Parts of the paper were presented at the Thirty-Ninth Asilomar Conference, Pacific Grove, CA, Oct. 30–November 2, 2005, and at the 2006 International Conference on Acoustics, Speech and Signal Processing (ICASSP), Toulouse, France, May 14–19, 2006.",

year = "2007",

month = aug,

doi = "10.1109/TSP.2007.895987",

language = "English (US)",

volume = "55",

pages = "4284--4299",

journal = "IEEE Transactions on Signal Processing",

issn = "1053-587X",

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

number = "8",

}

TY - JOUR

T1 - Distributed estimation using reduced-dimensionality sensor observations

AU - Schizas, Ioannis D.

AU - Giannakis, Georgios B.

AU - Luo, Zhi Quan

N1 - Funding Information: Manuscript received November 6, 2005; revised August 10, 2006. The associate editor coordinating the review of this manuscript and approving it for publication was Prof. Javier Garcia-Frias. Prepared through collaborative participation in the Communications and Networks Consortium sponsored by the U. S. Army Research Laboratory under the Collaborative Technology Alliance Program, Cooperative Agreement DAAD19-01-2-0011. The work of Z.-Q. Luo is supported by the U.S. DoD Army, grant number W911NF-05-1-0567. The U.S. Government is authorized to reproduce and distribute reprints for Government purposes notwithstanding any copyright notation thereon. Parts of the paper were presented at the Thirty-Ninth Asilomar Conference, Pacific Grove, CA, Oct. 30–November 2, 2005, and at the 2006 International Conference on Acoustics, Speech and Signal Processing (ICASSP), Toulouse, France, May 14–19, 2006.

PY - 2007/8

Y1 - 2007/8

N2 - We derive linear estimators of stationary random signals based on reduced-dimensionality observations collected at distributed sensors and communicated to a fusion center over wireless links. Dimensionality reduction compresses sensor data to meet low-power and bandwidth constraints, while linearity in compression and estimation are well motivated by the limited computing capabilities wireless sensor networks are envisioned to operate with, and by the desire to estimate random signals from observations with unknown probability density functions. In the absence of fading and fusion center noise (ideal links), we cast this intertwined compression-estimation problem in a canonical correlation analysis framework and derive closed-form mean-square error (MSE) optimal estimators along with coordinate descent suboptimal alternatives that guarantee convergence at least to a stationary point. Likewise, we develop estimators based on reduced-dimensionality sensor observations in the presence of fading and additive noise at the fusion center (nonideal links). Performance analysis and corroborating simulations demonstrate the merits of the novel distributed estimators relative to existing alternatives.

AB - We derive linear estimators of stationary random signals based on reduced-dimensionality observations collected at distributed sensors and communicated to a fusion center over wireless links. Dimensionality reduction compresses sensor data to meet low-power and bandwidth constraints, while linearity in compression and estimation are well motivated by the limited computing capabilities wireless sensor networks are envisioned to operate with, and by the desire to estimate random signals from observations with unknown probability density functions. In the absence of fading and fusion center noise (ideal links), we cast this intertwined compression-estimation problem in a canonical correlation analysis framework and derive closed-form mean-square error (MSE) optimal estimators along with coordinate descent suboptimal alternatives that guarantee convergence at least to a stationary point. Likewise, we develop estimators based on reduced-dimensionality sensor observations in the presence of fading and additive noise at the fusion center (nonideal links). Performance analysis and corroborating simulations demonstrate the merits of the novel distributed estimators relative to existing alternatives.

KW - Canonical correlation analysis (CCA)

KW - Distributed compression

KW - Distributed estimation

KW - Nonlinear optimization

KW - Wireless sensor networks (WSNs)

UR - http://www.scopus.com/inward/record.url?scp=34547899149&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34547899149&partnerID=8YFLogxK

U2 - 10.1109/TSP.2007.895987

DO - 10.1109/TSP.2007.895987

M3 - Article

AN - SCOPUS:34547899149

SN - 1053-587X

VL - 55

SP - 4284

EP - 4299

JO - IEEE Transactions on Signal Processing

JF - IEEE Transactions on Signal Processing

IS - 8

ER -

Distributed estimation using reduced-dimensionality sensor observations

Abstract

Bibliographical note

Keywords

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this