Error event simulation for HMM tracking algorithms using importance sampling

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

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Importance sampling is a technique for speeding up Monte Carlo (MC) simulations. The fundamental idea is to use a different simulation distribution to increase the relative frequency of important events and then weight the observed data in order to obtain an unbiased estimate of the parameter of interest. This estimate often requires orders-of-magnitude fewer simulation trials than ordinary MC simulations to obtain the same specified precision. In this paper, we present an importance sampling technique applicable to error event simulation of hidden Markov model (HMM) tracking algorithms. The computational savings possible with the use of this technique are demonstrated both analytically and using simulation results for a specific HMM tracking algorithm.

Original languageEnglish
Pages (from-to)720-736
Number of pages17
JournalIEEE Transactions on Signal Processing
Volume46
Issue number3
DOIs
Publication statusPublished - 1998
Externally publishedYes

Fingerprint

Importance sampling
Hidden Markov models
Monte Carlo simulation

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Error event simulation for HMM tracking algorithms using importance sampling. / Sanjeev Arulampalam, M.; Evans, Rob J.; Letaief, Khaled.

In: IEEE Transactions on Signal Processing, Vol. 46, No. 3, 1998, p. 720-736.

Research output: Contribution to journalArticle

@article{1cd9921c3cac40d18e44779327b35e65,
title = "Error event simulation for HMM tracking algorithms using importance sampling",
abstract = "Importance sampling is a technique for speeding up Monte Carlo (MC) simulations. The fundamental idea is to use a different simulation distribution to increase the relative frequency of important events and then weight the observed data in order to obtain an unbiased estimate of the parameter of interest. This estimate often requires orders-of-magnitude fewer simulation trials than ordinary MC simulations to obtain the same specified precision. In this paper, we present an importance sampling technique applicable to error event simulation of hidden Markov model (HMM) tracking algorithms. The computational savings possible with the use of this technique are demonstrated both analytically and using simulation results for a specific HMM tracking algorithm.",
author = "{Sanjeev Arulampalam}, M. and Evans, {Rob J.} and Khaled Letaief",
year = "1998",
doi = "10.1109/78.661338",
language = "English",
volume = "46",
pages = "720--736",
journal = "IEEE Transactions on Signal Processing",
issn = "1053-587X",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "3",

}

TY - JOUR

T1 - Error event simulation for HMM tracking algorithms using importance sampling

AU - Sanjeev Arulampalam, M.

AU - Evans, Rob J.

AU - Letaief, Khaled

PY - 1998

Y1 - 1998

N2 - Importance sampling is a technique for speeding up Monte Carlo (MC) simulations. The fundamental idea is to use a different simulation distribution to increase the relative frequency of important events and then weight the observed data in order to obtain an unbiased estimate of the parameter of interest. This estimate often requires orders-of-magnitude fewer simulation trials than ordinary MC simulations to obtain the same specified precision. In this paper, we present an importance sampling technique applicable to error event simulation of hidden Markov model (HMM) tracking algorithms. The computational savings possible with the use of this technique are demonstrated both analytically and using simulation results for a specific HMM tracking algorithm.

AB - Importance sampling is a technique for speeding up Monte Carlo (MC) simulations. The fundamental idea is to use a different simulation distribution to increase the relative frequency of important events and then weight the observed data in order to obtain an unbiased estimate of the parameter of interest. This estimate often requires orders-of-magnitude fewer simulation trials than ordinary MC simulations to obtain the same specified precision. In this paper, we present an importance sampling technique applicable to error event simulation of hidden Markov model (HMM) tracking algorithms. The computational savings possible with the use of this technique are demonstrated both analytically and using simulation results for a specific HMM tracking algorithm.

UR - http://www.scopus.com/inward/record.url?scp=0032028971&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032028971&partnerID=8YFLogxK

U2 - 10.1109/78.661338

DO - 10.1109/78.661338

M3 - Article

VL - 46

SP - 720

EP - 736

JO - IEEE Transactions on Signal Processing

JF - IEEE Transactions on Signal Processing

SN - 1053-587X

IS - 3

ER -