Computing R.O.C. for Quadratic Detectors.
NAVAL SURFACE WEAPONS CENTER WHITE OAK LAB SILVER SPRING MD
Pagination or Media Count:
After defining a broad class of nonlinear detectors--those whose outputs are quadratic forms in terms of their inputs, exact probability distributions are derived when time samples are independent for three important sub-classes coherent power sum and square, correlator, and incoherent power square and sum detectors. An asymptotic method is used to obtain the distributions of the more general class of detector, including dependent time samples. Computer programs, example calculations, and many graphical results are included the emphasis is on imparting a methodology which has been successful in computing R.O.C. for the detectors modeled, for both Gaussian and deterministic signals. Author
- Statistics and Probability
- Active and Passive Radar Detection and Equipment