Filtering With Two Sided Filtrations

Abstract

The value of a diffusion at an intermediate point is observed through noisy observations on each side. Corresponding semimartingale decompositions and recursive filtering equations are obtained. Such problems possibly arise in reconstructing images from noisy data if one wishes to estimate the signal at linear location 1.2 based on observations from 0 to t and 1-t to 1 either side. This estimation involves the two delta-fields, Yt, increasing in the positive t direction, and Z1-t increasing in the negative t direction.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 1990
Accession Number
ADA238498

Entities

People

  • Robert J. Elliott

Organizations

  • University of Alberta

Tags

Communities of Interest

  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Abstracts
  • Brownian Motion
  • Computing-Related Activities
  • Data Science
  • Decomposition
  • Diffusion
  • Equations
  • Filtration
  • Information Science
  • Integrals
  • Interdisciplinary Science
  • Markov Processes
  • Military Research
  • Observation
  • Probability
  • Random Variables
  • Statistics

Readers

  • Adaptive Control and Estimation with Uncertainty in Dynamic Systems.
  • Statistical inference.