An introduction to signal detection and estimation (2nd ed.)
An introduction to signal detection and estimation (2nd ed.)
On the existence of Gaussian noise (signal modelling)
ICASSP '91 Proceedings of the Acoustics, Speech, and Signal Processing, 1991. ICASSP-91., 1991 International Conference
| Hi-index | 0.00 |
Emerging from our general geometric theory for detection problems is the geometric structure of the space of probability measures. On this non-Riemannian manifold, distance cannot be defined, making it impossible to construct directly signal constellations for non-Gaussian detection problems. As the number of observations increases, an approximate structure for signal constellations can be obtained. We show that square-wave signals play a prominent role in non-Gaussian detection problems. A procedure for determining optimum signal sets is described. Optimum signal constellations depend on signal-to-noise ratio, in some cares changing their geometric form as signal-to-noise ratio changes