Skip to Main Content
This paper is an attempt to present a short, unified discussion of the radar detection, parameter estimation, and multiple-signal resolution problems--mostly from a philosophical rather than a detailed mathematical point of view. The purpose is, hopefully, to make it possible in at least some limited sense to reason back from appropriate measures of desired radar performance to specifications of the necessary values of the related radar parameters. Specifically four measures of performance quality are considered: 1.The reliability of detection, 2.The accuracy with which target parameters can be estimated, 3.The extent to which such estimates can be made without ambiguity, 4.The degree to which two or more different target echoes can be separated or resolved. It is argued that the radar synthesis problem can be split into two more-or-less independent phases. First, adjust such parameters as those appearing in the radar equation so that the received signal energy is sufficiently large for the degree of reliability of detection desired. The required value of energy is almost entirely independent of the character of the received echo signal waveform. The second phase is, then, to select the waveform in such a way that accuracy, ambiguity, and resolution requirements are met. The limitations on what can be achieved in terms of these three quality measures are discussed in relation to an uncertainty principle. For purposes of illustration several novel waveforms having unusual and useful properties are described.