Validating the Modeling and Simulation of a Generic Tracking Radar

Abstract

This report proposes acceptability criteria for validating the modeling and simulation of a generic tracking radar. The approach is based on a statistical hypothesis test that strives to minimize risk to both the model user and the model maker. The validation process is limited to the comparison of a set of Monte Carlo realizations of judiciously selected validation metrics with single discrete-event observations made by the actual sensor. The effectiveness of the criteria is examined with controlled numerical experiments whereby the impact of poor models for target signature and signal propagation effects on the simulation of the sensor's tracking function is explored. Results are summarized in a scorecard containing a list of rejection indices and rejection thresholds for the different validation metrics. The rejection thresholds take into account the effect of any statistical correlations present in individual validation metrics. Due to the unavailability of the probability density function of the observed behavior, which prevents the computation of the model user's risk directly, a family of normalized rejection thresholds, corresponding to different values of the model maker's risk, are included. Scorecards also reveal any cross-correlations that exist among select validation metrics. This feature of the scorecard can serve as a diagnostic tool-thus, aiding in modeling and simulation improvement.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jul 28, 2009
Accession Number
ADA502982

Entities

People

  • A. S. Brewster
  • H. C. Lambert
  • K. Dunn
  • S. R. Vogel

Organizations

  • Massachusetts Institute of Technology

Tags

Communities of Interest

  • Materials and Manufacturing Processes
  • Sensors
  • Weapons Technologies

DTIC Thesaurus Topics

  • Acceptability
  • Computations
  • Cross Correlation
  • Data Science
  • Detectors
  • Distribution Functions
  • Environment
  • Fire Control Radar
  • Information Science
  • Kalman Filters
  • Markov Processes
  • Probability
  • Probability Density Functions
  • Probability Distributions
  • Radar
  • Random Variables
  • Simulations

Readers

  • Computational Modeling and Simulation
  • Radar Systems Engineering.
  • Statistical inference.