Detection Performance of a Suboptimum Processor in Non-Gaussian Noise

Abstract

The performance of a suboptimum processor for three different signal types in three different noise backgrounds has been investigated numerically in terms of the receiver operating characteristics and compared with the performance of the optimum processor in the same environments. The three distributions considered are Gaussian, sech, and fourth power; the sech distribution decays exponentially for large arguments, while the fourth-power distribution decays inversely proportional to the fourth power. Noise with the fourth-power distribution is very heavy-tailed and constitutes a debilitating effect on standard linear processors in weak signal environments. The suboptimum processor consists of a hard limiter followed by an energy detector. In the presence of heavy-tailed noise Interference, this suboptimum processor approach loses on the order of 1 dB In signal detectability relative to the optimum processor. Lack of the detailed signal structure is not a detriment to this processor; furthermore, the information required to realize the absolute optimum processor will not likely be available in practice.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Apr 19, 2004
Accession Number
ADA423872

Entities

People

  • Albert H. Nuttall

Organizations

  • Naval Undersea Warfare Center

Tags

Communities of Interest

  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Additives (Chemicals)
  • Background Noise
  • Covariance
  • Detection
  • Detectors
  • Distribution Functions
  • Environment
  • False Alarms
  • Gaussian Noise
  • Noise
  • Probability
  • Probability Density Functions
  • Random Variables
  • Standards
  • Statistics
  • Undersea Warfare
  • Warning Systems

Fields of Study

  • Engineering
  • Physics

Readers

  • Atmospheric Science / Meteorology, specifically Wind Wave Turbulence.
  • Mathematical Modeling and Probability Theory.
  • Radio communications and signal processing.