On Sequential Detection in Dependent Noise.
ILLINOIS UNIV AT URBANA COORDINATED SCIENCE LAB
Pagination or Media Count:
The problem of designing memoryless sequential detection systems for constant signals in m-dependent noise processes is considered. Particular attention is given to the problem of antipodal signaling in an additive noise channel. Applying the criterion of minimum average sample size, the asymptotically optimal detector is shown to be characterized by the solution to a Fredholm integral equation. A paradox arising from the application of standard asymptotic analysis to this detection system is also considered. In addition, a numerical method and resulting solution to the integral equation or the Gaussian case is also given. Author
- Numerical Mathematics