Consiglio Nazionale delle Ricerche

Tipo di prodottoArticolo in rivista
TitoloDistributed Estimation From Relative Measurements of Heterogeneous and Uncertain Quality
Anno di pubblicazione2019
Formato
  • Elettronico
  • Cartaceo
Autore/iRavazzi, Chiara; Chan, Nelson P. K.; Frasca, Paolo
Affiliazioni autoriNational Research Council of Italy CNR; Univ. Groningen; CNRS, Inria, Grenoble INP, GIPSA-lab
Autori CNR e affiliazioni
  • PAOLO FRASCA
  • CHIARA RAVAZZI
Lingua/e
  • inglese
AbstractThis paper studies the problem of estimation from relative measurements in a graph, in which a vector indexed over the nodes has to be reconstructed from pairwise measurements of differences between its components associated with nodes connected by an edge. In order to model heterogeneity and uncertainty of the measurements, we assume them to be affected by additive noise distributed according to a Gaussian mixture. In this original setup, we formulate the problem of computing the maximum-likelihood estimates and we design two novel algorithms, based on least squares (LS) regression and expectation maximization (EM). The first algorithm (LS-EM) is centralized and performs the estimation from relative measurements, the soft classification of the measurements, and the estimation of the noise parameters. The second algorithm (Distributed LS-EM) is distributed and performs estimation and soft classification of the measurements, but requires the knowledge of the noise parameters. We provide rigorous proofs of convergence for both algorithms and we present numerical experiments to evaluate their performance and compare it with solutions from the literature. The experiments show the robustness of the proposed methods against different kinds of noise and, for the Distributed LS-EM, against errors in the knowledge of noise parameters.
Lingua abstractinglese
Altro abstract-
Lingua altro abstractinglese
Pagine da203
Pagine a217
Pagine totali15
RivistaIEEE transactions on signal and information processing over networks
Attiva dal 2015
Editore: Institute of Electrical and Electronics Engineers - Piscataway
Paese di pubblicazione: Stati Uniti d'America
Lingua: inglese
ISSN: 2373-776X
Titolo chiave: IEEE transactions on signal and information processing over networks
Titolo proprio: IEEE transactions on signal and information processing over networks
Titolo abbreviato: IEEE trans. signal inf. process. netw.
Numero volume della rivista5
Fascicolo della rivista2
DOI10.1109/TSIPN.2018.2869117
Verificato da refereeSì: Internazionale
Stato della pubblicazionePreprint
Indicizzazione (in banche dati controllate)
  • ISI Web of Science (WOS) (Codice:000467571500001)
Parole chiaveClassification, estimation theory, gaussian mixture models, maximum-likelihood estimation, sensor networks
Link (URL, URI)-
Titolo parallelo-
Data di accettazione-
Note/Altre informazioni-
Strutture CNR
  • IEIIT — Istituto di elettronica e di ingegneria dell'informazione e delle telecomunicazioni
Moduli/Attività/Sottoprogetti CNR
  • DIT.AD007.007.001 : CONES - Control of Networked Complex Systems
Progetti Europei-
Allegati
Preprint_TSIPN19
Tipo documento: application/pdf