Statistical Techniques for Determining the Repeatability of Man-in-the-Loop System Performance Data
Abstract
The Joint Advanced Distributed Simulation (JADS) Joint Test Force (JTF) was chartered by the Office of the Secretary of Defense to investigate the utility of advanced distributed simulation (ADS) technology to test and evaluation (T&E). JADS executed three test programs (command, control, communications, computers and intelligence (C4I); precision guided munitions; and electronic warfare) representing slices of the overall T&E spectrum, as well as observing other activity within the T&E community, to form its conclusions. One of the slices, the Electronic Warfare (EW) Test compared system performance data collected using ADS to data collected in a stand-alone hardware-in-the-loop (HITL) facility and on an open air range (OAR) via traditional test methods. In order to isolate and evaluate the impacts of distribution from other potential variance sources, testing was conducted under tightly controlled conditions. Yet, because of the statistical nature of the performance of man-in-the-loop systems, noise in electronic systems, and difficulty in replicating human and environmental conditions, it was extremely difficult to repeat a particular test without obtaining slightly different results. This paper discusses the application of statistical analysis techniques to evaluate the repeatability, or consistency, of the EW Test data collected within individual test facilities, in order to aid in the identification and understanding of the impacts of the variance sources within, and across, test phases.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 2000
- Accession Number
- ADA378561
Entities
People
- Darrell L. Wright
- Roman Nation
- Sandra K. Smith