A Novel Differential Geometric Approach toward Robust Signal Detection.

Abstract

We present a new approach toward robust signal detection which is based on techniques rooted in differential geometry. These methods, as opposed to the commonly employed classical saddlepoint criteria, readily admit the quantitive measure of the degree of robustness over very general classes of admissible noise distributions. Our approach thus is seen to make possible investigations of the quantative tradeoff between optimal performance and robustness, and we illustrate the application of this differential geometric approach via various specific examples. (Author)

Open PDF

Document Details

Document Type
Technical Report
Publication Date
May 22, 1986
Accession Number
ADA170035

Entities

People

  • D. R. Halverson
  • M. W. Thompson

Organizations

  • Texas A&M University

Tags

DTIC Thesaurus Topics

  • Classification
  • Coordinate Systems
  • Curvature
  • Detection
  • Detectors
  • Differential Geometry
  • Electrical Engineering
  • False Alarms
  • Geometry
  • New York
  • Security
  • Signal Detection
  • Step Functions
  • Two Dimensional
  • Universities
  • Warning Systems

Readers

  • Calculus or Mathematical Analysis
  • Statistical inference.
  • Systems Analysis and Design