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.
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