On the asymptotic scalability of the consensus algorithm

Mehmet E. Yildiz, Anna Scaglione

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

2 Citations (Scopus)

Abstract

Average consensus algorithms are gossiping protocols for averaging original measurements taken at different sensors. Without any communication rate restrictions, the algorithms ideally allow every node state to converge to the initial average after some iterations. Noting that brute force quantization is highly suboptimal given the rich temporal and spatial correlation of the messages exchanged, in our previous work we proposed two source coding methods, predictive coding and Wyner-Ziv coding which achieve convergence with vanishing quantization rates in the case of block coding. Both methods ideally require complete information about the network parameters and topology as well as processing and storage of the past state values. The knowledge of the network parameters is not a practical requirement, especially considering that one is implementing average consensus algorithms that are decentralized. In this study we show that as the node density increases or in homogeneously distributed networks, the encoder and decoder parameters become independent on the network size and specific location. We also lower bound the error performance of the predictive coding scheme in terms of eigen-values of the connectivity and initial state covariance matrices. We show the relation between MSE behavior of the algorithm and network connectivity, as well as quantization rate.

Original languageEnglish (US)
Title of host publicationIEEE Workshop on Statistical Signal Processing Proceedings
Pages645-649
Number of pages5
DOIs
StatePublished - 2007
Externally publishedYes
Event2007 IEEE/SP 14th WorkShoP on Statistical Signal Processing, SSP 2007 - Madison, WI, United States
Duration: Aug 26 2007Aug 29 2007

Other

Other2007 IEEE/SP 14th WorkShoP on Statistical Signal Processing, SSP 2007
CountryUnited States
CityMadison, WI
Period8/26/078/29/07

Fingerprint

Scalability
Covariance matrix
Topology
Communication
Sensors
Processing

Keywords

  • Communication systems
  • Distributed algorithms
  • Networks

ASJC Scopus subject areas

  • Signal Processing

Cite this

Yildiz, M. E., & Scaglione, A. (2007). On the asymptotic scalability of the consensus algorithm. In IEEE Workshop on Statistical Signal Processing Proceedings (pp. 645-649). [4301338] https://doi.org/10.1109/SSP.2007.4301338

On the asymptotic scalability of the consensus algorithm. / Yildiz, Mehmet E.; Scaglione, Anna.

IEEE Workshop on Statistical Signal Processing Proceedings. 2007. p. 645-649 4301338.

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

Yildiz, ME & Scaglione, A 2007, On the asymptotic scalability of the consensus algorithm. in IEEE Workshop on Statistical Signal Processing Proceedings., 4301338, pp. 645-649, 2007 IEEE/SP 14th WorkShoP on Statistical Signal Processing, SSP 2007, Madison, WI, United States, 8/26/07. https://doi.org/10.1109/SSP.2007.4301338
Yildiz ME, Scaglione A. On the asymptotic scalability of the consensus algorithm. In IEEE Workshop on Statistical Signal Processing Proceedings. 2007. p. 645-649. 4301338 https://doi.org/10.1109/SSP.2007.4301338
Yildiz, Mehmet E. ; Scaglione, Anna. / On the asymptotic scalability of the consensus algorithm. IEEE Workshop on Statistical Signal Processing Proceedings. 2007. pp. 645-649
@inproceedings{4143067ea23d43c987893c9f4cbd2b9a,
title = "On the asymptotic scalability of the consensus algorithm",
abstract = "Average consensus algorithms are gossiping protocols for averaging original measurements taken at different sensors. Without any communication rate restrictions, the algorithms ideally allow every node state to converge to the initial average after some iterations. Noting that brute force quantization is highly suboptimal given the rich temporal and spatial correlation of the messages exchanged, in our previous work we proposed two source coding methods, predictive coding and Wyner-Ziv coding which achieve convergence with vanishing quantization rates in the case of block coding. Both methods ideally require complete information about the network parameters and topology as well as processing and storage of the past state values. The knowledge of the network parameters is not a practical requirement, especially considering that one is implementing average consensus algorithms that are decentralized. In this study we show that as the node density increases or in homogeneously distributed networks, the encoder and decoder parameters become independent on the network size and specific location. We also lower bound the error performance of the predictive coding scheme in terms of eigen-values of the connectivity and initial state covariance matrices. We show the relation between MSE behavior of the algorithm and network connectivity, as well as quantization rate.",
keywords = "Communication systems, Distributed algorithms, Networks",
author = "Yildiz, {Mehmet E.} and Anna Scaglione",
year = "2007",
doi = "10.1109/SSP.2007.4301338",
language = "English (US)",
isbn = "142441198X",
pages = "645--649",
booktitle = "IEEE Workshop on Statistical Signal Processing Proceedings",

}

TY - GEN

T1 - On the asymptotic scalability of the consensus algorithm

AU - Yildiz, Mehmet E.

AU - Scaglione, Anna

PY - 2007

Y1 - 2007

N2 - Average consensus algorithms are gossiping protocols for averaging original measurements taken at different sensors. Without any communication rate restrictions, the algorithms ideally allow every node state to converge to the initial average after some iterations. Noting that brute force quantization is highly suboptimal given the rich temporal and spatial correlation of the messages exchanged, in our previous work we proposed two source coding methods, predictive coding and Wyner-Ziv coding which achieve convergence with vanishing quantization rates in the case of block coding. Both methods ideally require complete information about the network parameters and topology as well as processing and storage of the past state values. The knowledge of the network parameters is not a practical requirement, especially considering that one is implementing average consensus algorithms that are decentralized. In this study we show that as the node density increases or in homogeneously distributed networks, the encoder and decoder parameters become independent on the network size and specific location. We also lower bound the error performance of the predictive coding scheme in terms of eigen-values of the connectivity and initial state covariance matrices. We show the relation between MSE behavior of the algorithm and network connectivity, as well as quantization rate.

AB - Average consensus algorithms are gossiping protocols for averaging original measurements taken at different sensors. Without any communication rate restrictions, the algorithms ideally allow every node state to converge to the initial average after some iterations. Noting that brute force quantization is highly suboptimal given the rich temporal and spatial correlation of the messages exchanged, in our previous work we proposed two source coding methods, predictive coding and Wyner-Ziv coding which achieve convergence with vanishing quantization rates in the case of block coding. Both methods ideally require complete information about the network parameters and topology as well as processing and storage of the past state values. The knowledge of the network parameters is not a practical requirement, especially considering that one is implementing average consensus algorithms that are decentralized. In this study we show that as the node density increases or in homogeneously distributed networks, the encoder and decoder parameters become independent on the network size and specific location. We also lower bound the error performance of the predictive coding scheme in terms of eigen-values of the connectivity and initial state covariance matrices. We show the relation between MSE behavior of the algorithm and network connectivity, as well as quantization rate.

KW - Communication systems

KW - Distributed algorithms

KW - Networks

UR - http://www.scopus.com/inward/record.url?scp=47849100425&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=47849100425&partnerID=8YFLogxK

U2 - 10.1109/SSP.2007.4301338

DO - 10.1109/SSP.2007.4301338

M3 - Conference contribution

AN - SCOPUS:47849100425

SN - 142441198X

SN - 9781424411986

SP - 645

EP - 649

BT - IEEE Workshop on Statistical Signal Processing Proceedings

ER -