Statistically/Computationally Efficient Detection in Incompletely Characterized Colored Non-Gaussian Noise via Parametric Modeling

Abstract

A generalized likelihood ratio test is known to be able to reliably detect a signal known except for amplitude incompletely characterized colored non-gaussian noise, although it is computationally intensive. A Rao efficient score test shares all the asymptotic properties of the generalized likelihood ratio test for large data records and small signal amplitudes. Its detection performance is asymptotically equivalent to that obtained for a similar detector designed with a priori knowledge of the unknown noise parameters. Computer simulations of the performance of the Rao detector support the theoretical results. A Rao detector built with the knowledge of the true form of the noise PDF is shown to significantly outperform a detector which assumes the noise to be Gaussian.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Aug 01, 1986
Accession Number
ADA175508

Entities

People

  • Debasis Sengupta
  • Steven Kay

Organizations

  • University of Rhode Island

Tags

Communities of Interest

  • Ground and Sea Platforms
  • Human Systems

DTIC Thesaurus Topics

  • Computational Complexity
  • Computational Science
  • Computations
  • Computer Simulations
  • Detection
  • Detectors
  • Estimators
  • False Alarms
  • Gaussian Noise
  • Gaussian Processes
  • Military Research
  • Probability
  • Random Variables
  • Rhode Island
  • Simulations
  • Statistics
  • Warning Systems

Fields of Study

  • Engineering

Readers

  • Distributed Systems and Data Platform Development
  • Statistical inference.