Coincidence Prodedures for the Detection of Known Signals in Statistically Undefined Background Noise.
Abstract
Several classes of coincidence procedures are considered for the detection of known signals in additive signal-independent noise about which little statistical information is available, or can easily be obtained. The detection applications considered are of both the single channel and dual channel variety, the latter case including the situation where an independent reference channel is available as well as the dual channel test for the presence of a common-mode signal. The coincidence procedures under study are direct outgrowths of conventional optimum linear and/or quadratic filtering theory suggested by the robust performance of coincidence procedures in general. In each of the detection applications under consideration, optimum coincidence procedures are defined directly analogous to their analog counter parts.
Document Details
- Document Type
- Technical Report
- Publication Date
- Feb 01, 1975
- Accession Number
- ADA009433
Entities
People
- James W. Modestino
Organizations
- Rensselaer Polytechnic Institute