Abstract
According to studies done by M. I. Trifonov, the form of the optimal detection template for an observer model with additive internal noise depends not only on the signal to be detected, but also on the internal-external noise ratio. The optimal detection template described by Trifonov's theory indeed exhibited superior performance over the conventional "matched filter" template, but the mathematical procedure Trifonov proposed for proving his theory was insufficient. During the course of this thesis project, an extensive research was carried out to study Trifonov's theory and its implications. As a result of this research, a new mathematical technique was developed to fully prove Trifonov's theory. My contributions presented in this thesis also include the extension of the study to two dimensional signals and the Monte Carlo simulation experiments which were conducted to verify the mathematical theory. The experiment results were found to be in agreement with Trifonov's theory.
Library of Congress Subject Headings
Signal detection--Mathematical models; Signal processing--Mathematical models; Random noise theory
Publication Date
4-1-1996
Document Type
Thesis
Department, Program, or Center
Chester F. Carlson Center for Imaging Science (COS)
Advisor
Burgess, Arthur
Advisor/Committee Member
Dougherty, Edward
Advisor/Committee Member
Fairchild, Mark
Recommended Citation
Xue, Shan, "Optimization of template selection for signal detection systems with additive internal noise" (1996). Thesis. Rochester Institute of Technology. Accessed from
https://repository.rit.edu/theses/2832
Campus
RIT – Main Campus
Comments
Note: imported from RIT’s Digital Media Library running on DSpace to RIT Scholar Works. Physical copy available through RIT's The Wallace Library at: TK5102.9 .X84 1996