THE OUTPUT PROBABILITY DISTRIBUTION OF A CORRELATION DETECTOR WITH SIGNAL PLUS NOISE INPUTS
Abstract
The problem is studied of determining the probability density function of the output of a correlation detector whose two inputs consist of correlated signal corrupted by uncorrelated noise. The inputs are stationary and Gaussian, one having the characteristics of white noise and the other being RC-filtered white noise. The post-multiplier averager is also an RC filter. The general case of signal plus noise inputs is investigated; the special cases of signal-only and noise-only inputs are also considered. Detailed results are presented when the ratio of the time-constant of the post-multiplier filter to that of the pre-multiplier filter is 1/2 and also in the practical case when this ratio is large. With the probability distribution determined, a statistical theory of signal detection is applied, a major result being the determination of the relationship between detection probability and the classical detection measure, output signal-to-noise ratio. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 01, 1962
- Accession Number
- AD0276364
Entities
People
- Melvin J. Jacobson
Organizations
- Rensselaer Polytechnic Institute