Results on the Detection of Signals in Spherically Invariant Random Noise

Abstract

A relationship between well-known estimator-correlator results from detection theory and the detection of signals in spherically invariant random noise is described. This relationship gives a general result concerning the structure of the likelihood ratio for this detection problem. Furthermore, an alternate formulation of both the likelihood ratio and the optimal estimator that arises in the estimator-correlator structure is given. This alternate formulation is important because it yields a closed-form solution for this optimal estimator without requiring explicit knowledge of a prior distribution for the unknown quantity being estimated. Since interest in modeling actual noise processes as spherically invariant random processes has recently appeared, the results here should help not only to give insight into the optimal detection structure in such noise but also to give guidance in formulating suboptimal detectors for problems of practical interest.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Nov 17, 1989
Accession Number
ADA214933

Entities

People

  • Karl R. Gerlach
  • Kevin J. Sangston

Organizations

  • United States Naval Research Laboratory

Tags

DTIC Thesaurus Topics

  • Abstracts
  • Amplitude
  • Classification
  • Coordinate Systems
  • Correlators
  • Detection
  • Detectors
  • Estimators
  • False Alarms
  • Gaussian Distributions
  • Gaussian Processes
  • Matched Filters
  • Military Research
  • Optimal Estimators
  • Signal Detection
  • Signal Processing
  • Warning Systems

Readers

  • Adaptive Control and Estimation with Uncertainty in Dynamic Systems.
  • Approximation Theory.