Applications of Correlation Techniques for Battlefield Identification I.
Abstract
This study is the first in a series of reports involved in researching self-correlation and cross-correlation algorithms in intelligence systems. These algorithms are used to maintain a data base of current information about a battlefield. The initial view of the battlefield is stored in a central computer data base. As new data is received from the sensors on the battlefield, it is used to update the old data and formulate a new picture of the battlefield. The work on these algorithms reported here focuses on the sensitivity of the mathematical tests to changes and uncertainties in the data. The self-correlation algorithms use multivariate statistical tests to determine the equality of mean vectors from two different datasets. The statistical tests developed were variations of Hotelling's T sub 2-statistics. The main results deal with the analysis of the robustness of these statistics with respect to normality and equal covariance matrices. Additional keywords: Multivariate distributions; Multivariate skewness; Chi square tests; Mathematical models; Computerized Simulation.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 30, 1984
- Accession Number
- ADA155239
Entities
People
- C. Black
- D. E. Hochman
- F. Ghobadian
- J. Fiskin
- R. Clough
Organizations
- Jet Propulsion Laboratory