Fusion of Correlated Decisions Using Regular Vine Copulas

Shan Zhang, Lakshmi Narasimhan Theagarajan, Sora Choi, Pramod Kumar Varshney

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

In this paper, we propose a regular vine copula based methodology for the fusion of correlated decisions. Regular vine copula is an extremely flexible and powerful graphical model to characterize complex dependence among multiple modalities. It can express a multivariate copula by using a cascade of bivariate copulas, the so-called pair copulas. Assuming that local detectors are single threshold binary quantizers and taking complex dependence among sensor decisions into account, we design an optimal fusion rule using a regular vine copula under the Neyman-Pearson framework. In order to reduce the computational complexity resulting from the complex dependence, we propose an efficient and computationally light regular vine copula based optimal fusion algorithm. Numerical experiments are conducted to demonstrate the effectiveness of our approach.

Original languageEnglish (US)
Article number8651356
Pages (from-to)2066-2079
Number of pages14
JournalIEEE Transactions on Signal Processing
Volume67
Issue number8
DOIs
StatePublished - Apr 15 2019

Fingerprint

Fusion reactions
Computational complexity
Detectors
Sensors
Experiments
Optimal design

Keywords

  • decision fusion
  • dependence modeling
  • Distributed detection
  • regular vine copula
  • sensor fusion

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Fusion of Correlated Decisions Using Regular Vine Copulas. / Zhang, Shan; Theagarajan, Lakshmi Narasimhan; Choi, Sora; Varshney, Pramod Kumar.

In: IEEE Transactions on Signal Processing, Vol. 67, No. 8, 8651356, 15.04.2019, p. 2066-2079.

Research output: Contribution to journalArticle

Zhang, Shan ; Theagarajan, Lakshmi Narasimhan ; Choi, Sora ; Varshney, Pramod Kumar. / Fusion of Correlated Decisions Using Regular Vine Copulas. In: IEEE Transactions on Signal Processing. 2019 ; Vol. 67, No. 8. pp. 2066-2079.
@article{02bb016013be41c5b99addb2e05f52bd,
title = "Fusion of Correlated Decisions Using Regular Vine Copulas",
abstract = "In this paper, we propose a regular vine copula based methodology for the fusion of correlated decisions. Regular vine copula is an extremely flexible and powerful graphical model to characterize complex dependence among multiple modalities. It can express a multivariate copula by using a cascade of bivariate copulas, the so-called pair copulas. Assuming that local detectors are single threshold binary quantizers and taking complex dependence among sensor decisions into account, we design an optimal fusion rule using a regular vine copula under the Neyman-Pearson framework. In order to reduce the computational complexity resulting from the complex dependence, we propose an efficient and computationally light regular vine copula based optimal fusion algorithm. Numerical experiments are conducted to demonstrate the effectiveness of our approach.",
keywords = "decision fusion, dependence modeling, Distributed detection, regular vine copula, sensor fusion",
author = "Shan Zhang and Theagarajan, {Lakshmi Narasimhan} and Sora Choi and Varshney, {Pramod Kumar}",
year = "2019",
month = "4",
day = "15",
doi = "10.1109/TSP.2019.2901379",
language = "English (US)",
volume = "67",
pages = "2066--2079",
journal = "IEEE Transactions on Signal Processing",
issn = "1053-587X",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "8",

}

TY - JOUR

T1 - Fusion of Correlated Decisions Using Regular Vine Copulas

AU - Zhang, Shan

AU - Theagarajan, Lakshmi Narasimhan

AU - Choi, Sora

AU - Varshney, Pramod Kumar

PY - 2019/4/15

Y1 - 2019/4/15

N2 - In this paper, we propose a regular vine copula based methodology for the fusion of correlated decisions. Regular vine copula is an extremely flexible and powerful graphical model to characterize complex dependence among multiple modalities. It can express a multivariate copula by using a cascade of bivariate copulas, the so-called pair copulas. Assuming that local detectors are single threshold binary quantizers and taking complex dependence among sensor decisions into account, we design an optimal fusion rule using a regular vine copula under the Neyman-Pearson framework. In order to reduce the computational complexity resulting from the complex dependence, we propose an efficient and computationally light regular vine copula based optimal fusion algorithm. Numerical experiments are conducted to demonstrate the effectiveness of our approach.

AB - In this paper, we propose a regular vine copula based methodology for the fusion of correlated decisions. Regular vine copula is an extremely flexible and powerful graphical model to characterize complex dependence among multiple modalities. It can express a multivariate copula by using a cascade of bivariate copulas, the so-called pair copulas. Assuming that local detectors are single threshold binary quantizers and taking complex dependence among sensor decisions into account, we design an optimal fusion rule using a regular vine copula under the Neyman-Pearson framework. In order to reduce the computational complexity resulting from the complex dependence, we propose an efficient and computationally light regular vine copula based optimal fusion algorithm. Numerical experiments are conducted to demonstrate the effectiveness of our approach.

KW - decision fusion

KW - dependence modeling

KW - Distributed detection

KW - regular vine copula

KW - sensor fusion

UR - http://www.scopus.com/inward/record.url?scp=85062705023&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85062705023&partnerID=8YFLogxK

U2 - 10.1109/TSP.2019.2901379

DO - 10.1109/TSP.2019.2901379

M3 - Article

AN - SCOPUS:85062705023

VL - 67

SP - 2066

EP - 2079

JO - IEEE Transactions on Signal Processing

JF - IEEE Transactions on Signal Processing

SN - 1053-587X

IS - 8

M1 - 8651356

ER -