Detection Performance of Signals in Dependent Noise From a Gaussian Mixture Uncertainty Class.
Abstract
Detection performance in terms of probabilities of detection and false alarm (PD and PF) is derived for a desired signal embedded in dependent (correlated) multivariate noise from a Gaussian mixture uncertainty class. This uncertainty class is defined using upper and lower bounding functions on the univariate Gaussian mixing distribution function. For a given detector operating point, expressions for PD and PF are derived. In addition, nontrivial lower and upper bounds of performance are derived which are functions of the bounding functions of the uncertainty class, the detection threshold, and the desired signal strength (only for PD).
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 31, 1998
- Accession Number
- ADA352456
Entities
People
- K. R. Gerlach
Organizations
- United States Naval Research Laboratory