Cooperative Localization for Mobile Networks

A Distributed Belief Propagation – Mean Field Message Passing Algorithm

Burak Cakmak, Daniel Nygaard Urup, Florian Meyer, Troels Pedersen, Bernard Henri Fleury, Franz Hlawatsch

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

9 Citationer (Scopus)
221 Downloads (Pure)

Resumé

We propose a hybrid message passing method
for distributed cooperative localization and tracking of mobile
agents. Belief propagation and mean field message passing are
employed for, respectively, the motion-related and measurementrelated
part of the factor graph. Using a Gaussian belief approximation,
only three real values per message passing iteration
have to be broadcast to neighboring agents. Despite these very
low communication requirements, the estimation accuracy can
be comparable to that of particle-based belief propagation.
OriginalsprogEngelsk
TidsskriftI E E E Signal Processing Letters
Vol/bind23
Udgave nummer6
Sider (fra-til)828 - 832
ISSN1070-9908
DOI
StatusUdgivet - 5 apr. 2016

Fingerprint

Message-passing Algorithms
Belief Propagation
Message passing
Mobile Networks
Message Passing
Mean Field
Wireless networks
Factor Graph
Mobile Agent
Broadcast
Motion
Communication
Requirements
Approximation

Citer dette

@article{fc56b9e1beae4b5da7cdc2e91ee358ee,
title = "Cooperative Localization for Mobile Networks: A Distributed Belief Propagation – Mean Field Message Passing Algorithm",
abstract = "We propose a hybrid message passing methodfor distributed cooperative localization and tracking of mobileagents. Belief propagation and mean field message passing areemployed for, respectively, the motion-related and measurementrelatedpart of the factor graph. Using a Gaussian belief approximation,only three real values per message passing iterationhave to be broadcast to neighboring agents. Despite these verylow communication requirements, the estimation accuracy canbe comparable to that of particle-based belief propagation.",
keywords = "Belief propagation , mean field approximation, cooperative localization, distributed estimation, information projection, Kullback-Leibler-divergence,, mobile agent network",
author = "Burak Cakmak and Urup, {Daniel Nygaard} and Florian Meyer and Troels Pedersen and Fleury, {Bernard Henri} and Franz Hlawatsch",
year = "2016",
month = "4",
day = "5",
doi = "10.1109/LSP.2016.2550534",
language = "English",
volume = "23",
pages = "828 -- 832",
journal = "I E E E Signal Processing Letters",
issn = "1070-9908",
publisher = "IEEE",
number = "6",

}

Cooperative Localization for Mobile Networks : A Distributed Belief Propagation – Mean Field Message Passing Algorithm. / Cakmak, Burak; Urup, Daniel Nygaard; Meyer, Florian; Pedersen, Troels; Fleury, Bernard Henri; Hlawatsch, Franz .

I: I E E E Signal Processing Letters, Bind 23, Nr. 6, 05.04.2016, s. 828 - 832.

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

TY - JOUR

T1 - Cooperative Localization for Mobile Networks

T2 - A Distributed Belief Propagation – Mean Field Message Passing Algorithm

AU - Cakmak, Burak

AU - Urup, Daniel Nygaard

AU - Meyer, Florian

AU - Pedersen, Troels

AU - Fleury, Bernard Henri

AU - Hlawatsch, Franz

PY - 2016/4/5

Y1 - 2016/4/5

N2 - We propose a hybrid message passing methodfor distributed cooperative localization and tracking of mobileagents. Belief propagation and mean field message passing areemployed for, respectively, the motion-related and measurementrelatedpart of the factor graph. Using a Gaussian belief approximation,only three real values per message passing iterationhave to be broadcast to neighboring agents. Despite these verylow communication requirements, the estimation accuracy canbe comparable to that of particle-based belief propagation.

AB - We propose a hybrid message passing methodfor distributed cooperative localization and tracking of mobileagents. Belief propagation and mean field message passing areemployed for, respectively, the motion-related and measurementrelatedpart of the factor graph. Using a Gaussian belief approximation,only three real values per message passing iterationhave to be broadcast to neighboring agents. Despite these verylow communication requirements, the estimation accuracy canbe comparable to that of particle-based belief propagation.

KW - Belief propagation

KW - mean field approximation

KW - cooperative localization

KW - distributed estimation

KW - information projection

KW - Kullback-Leibler-divergence,

KW - mobile agent network

U2 - 10.1109/LSP.2016.2550534

DO - 10.1109/LSP.2016.2550534

M3 - Journal article

VL - 23

SP - 828

EP - 832

JO - I E E E Signal Processing Letters

JF - I E E E Signal Processing Letters

SN - 1070-9908

IS - 6

ER -