Use of Distance versus Statistical Distance in Acceptance Tests
Abstract
The problem treated in this report is that of deciding, with a given (50%) confidence level and based on distance alone, whether two estimated fixes belong to the same emitter. In place of geometric distance - (x-y)T(x-y) - the authors propose use of statistical distance - (x-y)TB-1(x-y) - where B is a covariance matrix computed from those matrices of the fixes. The two methods are compared for various ratios of the axes of the 50% probability ellipse. Keywords: Covariance matrix, Efficiency of Test, Fix combination, Error ellipse.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 29, 1987
- Accession Number
- ADA197839
Entities
Organizations
- California Institute of Technology