Answers are grounded in this paper and cite their sources.
The paper deals with the problem, frequently encountered in radar, of extracting simple numerical information from a noisy waveform. It is suggested that the only ideal way of doing this is to use the principle of inverse probability and convert the wave-form into a probability distribution for the quantity sought. The method is applied to the problem of determining the time delay of a periodically modulated rf waveform in the presence of white Gaussian noise when the undelayed waveform without noise is exactly known. As a result, the matched predetection filter of Van Vleck and Middleton is automatically specified, and the theory of ideal detection is briefly indicated.
No takes yet. Share an insight, caveat, or question.
Philip Woodward (1951) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: