Importance sampling for error event analysis of HMM frequency line trackers

M. Sanjeev Arulampalam, Rob J. Evans, Khaled Letaief

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

This paper considers the problem of designing efficient and systematic importance sampling (IS) schemes for the performance study of hidden Markov model (HMM) based trackers. Importance sampling (IS) is a powerful Monte Carlo (MC) variance reduction technique, which can require orders of magnitude fewer simulation trials than ordinary MC to obtain the same specified precision. In this paper, we present an IS technique applicable to error event analysis of HMM based trackers. Specifically, we use conditional IS to extend our work in another of our papers to estimate average error event probabilities. In addition, we derive upper bounds on these error probabilities, which are then used to verify the simulations. The power and accuracy of the proposed method is illustrated by application to an HMM frequency tracker.

Original languageEnglish
Pages (from-to)411-424
Number of pages14
JournalIEEE Transactions on Signal Processing
Volume50
Issue number2
DOIs
Publication statusPublished - Feb 2002
Externally publishedYes

Fingerprint

Importance sampling
Hidden Markov models

Keywords

  • Error events
  • HMM
  • Importance sampling
  • Monte Carlo

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Importance sampling for error event analysis of HMM frequency line trackers. / Arulampalam, M. Sanjeev; Evans, Rob J.; Letaief, Khaled.

In: IEEE Transactions on Signal Processing, Vol. 50, No. 2, 02.2002, p. 411-424.

Research output: Contribution to journalArticle

@article{dca3d2e54415445090e62f2c4a26ce92,
title = "Importance sampling for error event analysis of HMM frequency line trackers",
abstract = "This paper considers the problem of designing efficient and systematic importance sampling (IS) schemes for the performance study of hidden Markov model (HMM) based trackers. Importance sampling (IS) is a powerful Monte Carlo (MC) variance reduction technique, which can require orders of magnitude fewer simulation trials than ordinary MC to obtain the same specified precision. In this paper, we present an IS technique applicable to error event analysis of HMM based trackers. Specifically, we use conditional IS to extend our work in another of our papers to estimate average error event probabilities. In addition, we derive upper bounds on these error probabilities, which are then used to verify the simulations. The power and accuracy of the proposed method is illustrated by application to an HMM frequency tracker.",
keywords = "Error events, HMM, Importance sampling, Monte Carlo",
author = "Arulampalam, {M. Sanjeev} and Evans, {Rob J.} and Khaled Letaief",
year = "2002",
month = "2",
doi = "10.1109/78.978395",
language = "English",
volume = "50",
pages = "411--424",
journal = "IEEE Transactions on Signal Processing",
issn = "1053-587X",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "2",

}

TY - JOUR

T1 - Importance sampling for error event analysis of HMM frequency line trackers

AU - Arulampalam, M. Sanjeev

AU - Evans, Rob J.

AU - Letaief, Khaled

PY - 2002/2

Y1 - 2002/2

N2 - This paper considers the problem of designing efficient and systematic importance sampling (IS) schemes for the performance study of hidden Markov model (HMM) based trackers. Importance sampling (IS) is a powerful Monte Carlo (MC) variance reduction technique, which can require orders of magnitude fewer simulation trials than ordinary MC to obtain the same specified precision. In this paper, we present an IS technique applicable to error event analysis of HMM based trackers. Specifically, we use conditional IS to extend our work in another of our papers to estimate average error event probabilities. In addition, we derive upper bounds on these error probabilities, which are then used to verify the simulations. The power and accuracy of the proposed method is illustrated by application to an HMM frequency tracker.

AB - This paper considers the problem of designing efficient and systematic importance sampling (IS) schemes for the performance study of hidden Markov model (HMM) based trackers. Importance sampling (IS) is a powerful Monte Carlo (MC) variance reduction technique, which can require orders of magnitude fewer simulation trials than ordinary MC to obtain the same specified precision. In this paper, we present an IS technique applicable to error event analysis of HMM based trackers. Specifically, we use conditional IS to extend our work in another of our papers to estimate average error event probabilities. In addition, we derive upper bounds on these error probabilities, which are then used to verify the simulations. The power and accuracy of the proposed method is illustrated by application to an HMM frequency tracker.

KW - Error events

KW - HMM

KW - Importance sampling

KW - Monte Carlo

UR - http://www.scopus.com/inward/record.url?scp=0036475041&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036475041&partnerID=8YFLogxK

U2 - 10.1109/78.978395

DO - 10.1109/78.978395

M3 - Article

VL - 50

SP - 411

EP - 424

JO - IEEE Transactions on Signal Processing

JF - IEEE Transactions on Signal Processing

SN - 1053-587X

IS - 2

ER -