Future Location Prediction using Hidden Markov Modeling

Abstract

Mobility of mobile SCUD launchers in Desert Storm produced a need for improved methods of tracking and location. Without ways to track SCUD launchers, great amounts of time and resources will continue to be wasted on search and destroy in future conflicts. Hidden Markov modeling provides a novel approach to predicting future movements of mobile SCUD launchers and other types of vehicles. The power of the hidden Markov model lies in the fact that it is a doubly stochastic process. With this process we can not only model the output of the system but also the model underlying Markov states that the system will visit. Knowing the underlying state transitions can supply an added benefit in predicting future movements.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Oct 26, 1998
Accession Number
ADA356326

Entities

People

  • Matthew A. Schnoor

Organizations

  • University of Colorado, at Colorado Springs

Tags

Communities of Interest

  • Materials and Manufacturing Processes
  • Space
  • Weapons Technologies

DTIC Thesaurus Topics

  • Air Force
  • Algorithms
  • Automated Speech Recognition
  • Computational Science
  • Computer Programming
  • Computers
  • Engineering
  • Hidden Markov Models
  • Markov Models
  • Markov Processes
  • New York
  • Operations Research
  • Probability
  • Probability Distributions
  • Random Variables
  • Stochastic Processes
  • Synthetic Aperture Radar

Readers

  • Computational Modeling and Simulation
  • Military History / Militaries and War Studies
  • Statistical inference.