Resolution limitations of a finite aperture

Abstract
Methods of increasing antenna resolution by data processing are of interest in radio astronomy and radiometric mapping applications. This paper deals with the fundamental limitations of such operations. Communication theory techniques using prolate spheroidal wave functions have already been applied to antenna pattern synthesis. These techniques are applied to the problem of reconstituting an observed temperature distribution. The manner in which noise limits this reconstitution process is clearly shown. A least expected mean square error reconstitution process is also detailed. An interesting analogy between data processing and supergaining is noted. Finally, the limitation set by noise on the resolution of a data processed finite aperture is concluded.