A Correlation Study of Some Factors Effecting Submarine Detection by Destroyer Mounted Sonars.
Abstract
A FORTRAN IV computer program was employed to conduct a statistical analysis of data collected during fleet antisubmarine warfare exercises. The object of the investigation was the identification of those variables which had greatest influence on a destroyer's ability to detect a submarine under certain conditions. The variables were treated as a random vector arising from one of two multivariate normal populations with common covariance matrix. An artificial regression relation was formulated to facilitate development of a linear discriminant function in a subset of those variables found to be of dominant importance. This latter subset was identified by examination of multiple correlation coefficients. The discriminant function was found to be seventy five per cent effective in classifying the experimental data correctly. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 1970
- Accession Number
- AD0722573
Entities
People
- Daniel Joseph Sullivan
Organizations
- Naval Postgraduate School