Performance Parameter Methodology Update for Information Delivery Times

Abstract

This report documents the development of a methodology for determining the overall information delivery time specifications for the Cheyenne Mountain Upgrade (CMU) from the individual subsystem transit time specifications. Since little is known about the transit time distribution of each subsystem, each is approximated by a rectangular distribution. Then these rectangular distributions are convolved to determine the distribution of the overall system performance parameter, information delivery time. When many rectangular distributions must be convolved, a simplification employing the Central Limit Theorem is used to estimate the mean and variance of the system's overall distribution. This distribution is thought to be close to normal with no negative tail. When rectangular distributions with large variance are used, errors in the resulting distribution can occur. These errors, introduced by the simplification technique, are reduced either by using numerical methods for convolving discrete distributions sampled from the continuous curve or, when less accuracy is required, by an approximation technique involving the convolution of two uniform distributions. A comparison of the numerical results derived from this approach with results calculated in a previous study is also presented.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 01, 1992
Accession Number
ADA248897

Entities

People

  • Barbara H. Roberts

Organizations

  • MITRE Corporation

Tags

Communities of Interest

  • Materials and Manufacturing Processes
  • Weapons Technologies

DTIC Thesaurus Topics

  • Acquisition
  • Air Force
  • Convolution
  • Engineering
  • Errors
  • Mathematics
  • Normal Distribution
  • Probability
  • Probability Density Functions
  • Random Variables
  • Specifications
  • Standards
  • Statistics
  • Stochastic Processes
  • Systems Engineering
  • Test And Evaluation
  • Warning Systems

Fields of Study

  • Mathematics

Readers

  • Software Engineering
  • Statistical inference.