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.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 1992
- Accession Number
- ADA248897
Entities
People
- Barbara H. Roberts
Organizations
- MITRE Corporation