A QUANTITATIVE EXAMINATION OF THE RADAR RESOLUTION PROBLEM
Abstract
The ability of a radar to resolve overlapping signals is examined. It is shown that the primary limitation on resolvability is the unpredictable difference between the actual received signal and the waveform for which the target receiver is matched (the distortion in the received signal). The effects of signal distortion on the shape of the 'ambiguity function' are examined, based upon a wideband random process model for the distortion. The statistics of the ambiguity function side-lobe region, in the presence of the distortion, are related in a simple manner to the statistics of the distortion process. The optimum two-target resolver for the distorted signal is then derived and its performance is examined for two different conditions of a priori knowledge concerning the signal: all parameters of both target returns known exactly, with only the presence or absence of the second target uncertain, and the phase and amplitude of either or both returns unknown, but all other parameters exactly known. These two cases make evident the comparative seriousness of wide-band distortion and lack of a priori knowledge. Probability of detection vs probability of false-alarm curves are derived for both cases.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 17, 1962
- Accession Number
- AD0295573
Entities
People
- J. L. Allen
Organizations
- Massachusetts Institute of Technology